PERFLUORINATED ALKYLATED SUBSTANCES (PFAS) IN THE NORDIC ENVIRONMENT Authors: Roland Kallenborn, Norwegian Institute for Air Research (NILU) Urs Berger, Norwegian Institute for Air Research (NILU) Ulf Järnberg, Institute for Applied Environmental Research (ITM), Stockholm University, Sweden Contributing authors: Maria Dam, Food, Veterinary and Environmental Agency of the Faroe Islands Ola Glesne, Norwegian Pollution Control Authority Britta Hedlund, Swedish Environmental Protection Agency Juha-Pekka Hirvi, Finnish Environment Institute Alf Lundgren, The Swedish Chemicals Inspectorate (representing the Nordic Chemicals Group). Betty Bügel Mogensen, National Environmental Research Institute of Denmark Albert S. Sigurdsson, Environment and Food Agency of Iceland
Transcript
PERFLUORINATED ALKYLATED SUBSTANCES (PFAS) IN THE NORDIC ENVIRONMENT
Authors: Roland Kallenborn, Norwegian Institute for Air Research (NILU)
Urs Berger, Norwegian Institute for Air Research (NILU)
Ulf Järnberg, Institute for Applied Environmental Research (ITM), Stockholm University, Sweden
Contributing authors: Maria Dam, Food, Veterinary and Environmental Agency of the Faroe Islands
Ola Glesne, Norwegian Pollution Control Authority
Britta Hedlund, Swedish Environmental Protection Agency
Juha-Pekka Hirvi, Finnish Environment Institute
Alf Lundgren, The Swedish Chemicals Inspectorate (representing the Nordic Chemicals Group).
Betty Bügel Mogensen, National Environmental Research Institute of Denmark
Albert S. Sigurdsson, Environment and Food Agency of Iceland
2
3
4
4
Table of contents
Table of contents.......................................................................................... 4 Preface ......................................................................................................... 7 Summary ...................................................................................................... 8 Sammendrag................................................................................................. 9 1. Frame of the study .............................................................................. 10 2. Background information ..................................................................... 12
2.1. Applications and product information ......................................... 14 3. Samples for PFAS screening in the Nordic environment .................... 15
3.3. Analysis and quantification.......................................................... 19 3.3.1. Preparation of sediment and sludge samples ..................... 20 3.3.2. Preparation of water samples............................................. 20 3.3.3. Preparation of biota samples.............................................. 21 3.3.4. Quantification of abiota samples........................................ 22 3.3.5. Quantification of biota samples ......................................... 24
3.4. Quality control and method comparison...................................... 24 3.4.1. Sampling protocol .............................................................. 25 3.4.2. Limit of detection/ limit of quantification.......................... 26 3.4.3. Laboratory and field blanks ............................................... 26 3.4.4. Performance tests (parallel samples) ................................. 28 3.4.5. Laboratory intercomparison............................................... 29 3.4.6. Data handling and comparison........................................... 30
4. Results................................................................................................. 31 4.1. Levels and distribution................................................................. 33
During the past decade, the identification of a new relevant group of environmental pollutants (i.e., perfluorinated alkylated substances = PFAS) has opened a new chapter within the various disciplines of environmental sciences. A relatively rapid development within the field of instrumental trace analysis, in combination with increased public environmental awareness has led to new concepts in detection, evaluation and remediation of potentially hazardous chemicals in the environment.
However, the group of perfluorinated alkylated substances (PFAS), recently marked as already ubiquitously distributed around our globe, demands new evaluation tools with regard to the technical detection as well as risk evaluation. The unique physico-chemical properties of the substance group are still a considerable challenge for environmental scientists as well as for regulatory authorities.
1.) The trace analysis of these compounds in environmental samples requires new highly sophisticated analytical instruments, such as high-performance liquid chromatography coupled to mass selective detectors in order to allow sufficient sensitive detections in the environment. Currently, these analytical methods, which are still under development, are suitable for research and method development but in most cases, not sufficiently validated for long-term monitoring in environmental compartments.
2.) PFAS residues are virtually both lipophobic and hydrophobic. In addition, these compounds express strong surface-active properties and usually adsorb strongly on natural surfaces. Thus, risk assessment tools, highly useful for conventional persistent organic pollutants (POPs), like the octanol-water partitioning coefficient (KOW) etc., are not suitable for evaluation of the environmental fate of PFAS residues.
3.) Due to the strong C-F binding in the PFAS molecule, most of these compounds are extremely persistent, virtually indestructible and are, thus, expected to prevail in the environment.
International regulatory authorities are currently discussing adequate measures to control and reduce the presence of PFAS related residues in the environment. The here presented report contribute to the world-wide efforts to evaluate and put in place suitable effective measures to reduce environmental hazards posed by PFOS and other relevant PFAS related chemicals.
The project team
8
8
Summary
The here presented screening study on occurrence, distribution and fate of perfluorinated alkylated substances (PFAS) and related chemicals in the Nordic environment revealed an ubiquitous distribution of perfluorinated contaminants. Six Nordic countries participated in the screening study (Denmark, Faroe Islands, Finland Iceland, Norway and Sweden). Compound specific distribution patterns in the different sample types confirmed that the physico-chemical properties in combination with release patterns and bioaccumulation potential are important and selective parameters for PFAS contamination.
High concentrations of PFAS related residues in sewage sludge and landfill effluents confirmed that these sample types are important primary anthropogenic sources for releases into the environment. Perfluorooctane sulfonate (PFOS) and perfluoroctanoic acid (PFOA) dominated in sewage sludge samples. Landfill effluent was highest contaminated of all aqueous samples. PFOA was dominating in landfill effluents. Lake water, seawater and rainwater (precipitation) samples were relatively low contaminated. However, measurable amounts of PFAS were found in all samples. The Nordic biota samples showed signals of species dependent distribution and levels. Highest PFAS levels were found in top predating Danish harbour seal (Phoca vitulina) samples with PFOS as predominant PFAS contaminant. However, in Faroe Island pilot whales (Globicephala melas), PFOSA and PFOS were dominating with up to 364 ng/g wet weight. Also Finnish and Norwegian pike samples (Esox lucius) are highly contaminated with PFOS (PFOS = 551 ng/g ww) demonstrating that also the freshwater ecosystem is contaminated with PFAS related chemicals. The patterns found in biota point towards both country specific release patterns and species depended up-take/ accumulation properties. The fact that PFOS and PFOSA were also detected in anadromous Arctic char in the Faroe Islands indicates that long-range transport in air and/or precipitation is occurring.
Thus, PFAS related chemicals are widely distributed in the Nordic environment. The presence of this type of compounds is generally confirmed for all environmental compartments. It is therefore recommended to include the relevant PFAS-related chemicals in environmental monitoring and consider further measures to reduce the burden of PFAS to the Nordic environment.
9 of 107
Sammendrag
En første nordisk ”screening”-undersøkelse om perfluorerte alkylerte stoffer (PFAS) med fokus på skjebne og fordeling av stoffene i miljøet, ble gjennomført med finansiell støtte av Nordisk ministerrådet (NMR). Seks nordiske land deltok i undersøkelsen (Denmark, Finland, Færøyene, Island, Norge og Sverige).
Komponent-spesifikke fordelingsmønstre i de ulike prøvetypene bekrefter at kjemiske egenskaper, i kombinasjon med utslippsmønster og akkumuleringspotensial er svært viktige parametre for opptak og spredning av PFAS i miljøet.
Høye nivåer av PFAS i kloakk og sigevannsprøver peker på at dette er viktige menneskeskapte primære kilder for utslipp og spredning av PFAS i det nordiske miljø. Perfluoroktansulfonat (PFOS) og perfluoroktansyre (PFOA) dominerer i kloakkprøvene fra alle seks nordiske land. I sigevannsprøvene (tatt fra lokaliteter i to nordiske land: Norge og Finland) dominerer PFOA. Sigevannsprøvene er betydelig høyere belastet med PFAS enn kloakkprøvene (opp til SUM PFAS = 1537 ng/g ww). Sammenlignet med kloakk og sigevann er PFAS nivåene i ferskvann (Mjøsa), saltvann og regnvann relativt lave. Men tilstedeværelsen av PFOS og PFOSA i anadrom røye fra Farøyene er en indikasjon for at atmosfærisk langtransport og nedbør er viktige kilder for PFAS i miljøet.
Indikasjoner for artspesifikke anrikningsmønster ble funnet i det biologiske materialet (16 ulike marine og ferskvann arter). Relativ høye PFAS-konsentrasjoner er påvist i marine og ferskvanns topp-predatorer. Høyest PFAS-belastning ble funnet i steinkobbe (Phoca vitulina) lever fra Danmark (PFOS = 551 ng/g ww), men også i grindhval (Globicephala melas) fra Færøyene var PFAS nivåene høyt (PWFAR09: PFOSA =364 ng/g ww). PFOS dominerte i selene, mens PFOSA var høyest konsentrert i grindhval-lever.
I ferskvannsfisk ble det funnet høye konsentrasjoner av PFAS i gjedde (Esox lucius). Konsentrasjoner i ferskvannsfisk var sammenlignbart med nivåene detektert i marine topp-predatorer (PFOS = 551 ng/g ww). Også i lever fra gjedde var PFOS den dominerende PFAS komponenten.
Basert på resultatene fremlagt i denne rapporten kan det konkluderes at PFAS også finnes i betydelige konsentrasjoner i det nordiske miljø. I alle prøvetyper som er analysert ble det funnet PFAS. Det anbefales derfor å inkludere PFAS i nasjonale overvåkingsprogrammer vurdere ytterlige tiltak som kan redusere belastningen fra PFAS i miljøet.
10
10
1. Frame of the study
A first screening project on the fate of perfluorinated alkylated substances (PFAS) in the Nordic environment was initiated by a project group with representatives of National Environmental Research Institute of Denmark, Finnish Environment Institute, Environment and Food Agency of Iceland, Food, Veterinary and Environmental Agency of the Faroe Islands, Norwegian Pollution Control Authority and Swedish Environmental Protection Agency. The project was financed and supported by the Nordic Council of Ministers through the Nordic Chemicals Group and the Nordic Monitoring and Data Group as well as the participating institutions.
The respective participating Nordic countries organized sample selection, collection and transport based on a sample protocol and manuals provided by the analytical laboratories. Sample preparation, trace analysis and quantification were performed jointly by the Norwegian Institute for Air Research (NILU, Kjeller, Norway) and the Institute for Applied Environmental Research (ITM, Stockholm, Sweden). For this first screening, six PFAS related chemicals with potential for accumulation in the environment were chosen. In addition, the analytical laboratories voluntarily added two PFAS compounds (PFBS and PFDS,) to the list of target chemicals (table 1).
The selected PFAS related compounds were chosen due to the following priority criteria:
- Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) are the two compounds best documented of all PFAS related compounds so far with regard to environmental hazards. PFOS and PFOA, thus, are the major motivation for the presented screening project.
- Perfluorooctane sulfonamide (PFOSA) is suspected to be a major precursor of PFOS in the environment.
- Perfluorobutane sulfonate (PFBS), perfluorohexane sulfonate (PFHxS) and perfluorodecane sulfonate (PFDS) are expected to be found in the environment due to their similar structure compared with PFOS. However only sparse information about environmental properties and toxicity is available yet. In addition, PFBS is announced as successor for PFOS-related products.
- Perfluorohexanoic acid (PFHxA), perfluoroheptanoic acid (PFHpA) and perfluorononanoic acid (PFNA) are expected to be present in the environment due to their resemblance with PFOA. However, only sparse information about environmental properties and toxicity is available yet.
For most of the substances included, indications existed already for their occurrence in the environment. The expected results will allow assessing the existing level of contamination (spatial distribution monitoring) and indicating regional differences. This spatial screening programme will enable the determination of the representativeness of the monitoring sites with regard to spatial variability in contaminant concentrations and will give valuable information about the ubiquity of PFAS distribution in the Nordic countries.
11 of 107
Table 1: Selected PFAS related chemicals chosen for the presented Nordic screening programme.
Acronym Name CAS-number Structure
PFBS Perfluorobutane sulfonate 29420-49-3 S O
O
F
F
FF
FF
F
FF
O
PFHxS Perfluorohexane sulfonate 432-50-7
F
S O
O
F
F
FF
FF
FF
FF
F
F
O
PFOS Perfluorooctane sulfonate 2795-39-3
S O
O
F
F
FF
FF
FF
FF
FF
FF
FF
F
O
PFDS Perfluorodecane sulfonate 67906-42-7
S
O
OO
F
FF
FF
FF
FF
F
FF
FF
FF
F
F
F
F
PFHxA Perfluorohexanoic acid 307-24-4
F
FF
FF
FF
FF
F
F
O
O
PFHpA Perfluoroheptanoic acid 375-85-9
F
FF
FF
FF
FF
FF
F F O
O
PFOA Perfluorooctanoic acid 335-67-1
FF
FF
FF
FF
FF
F
F
FF O
O
F
PFNA Perfluorononanoic acid 375-95-1
FF
FF
FF
FF
FF
F
F
FF O
O
F
F
F
PFOSA Perfluorooctane
sulfonamide 4151-50-2
S N
O
F
F
FF
FF
FF
FF
FF
F
F
FF
H
H
F
O
12
12
2. Background information
Within the past 50 years, industrial application as well as consumer uses of perfluorinated alkylated substances (PFAS) such as perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA), as well as their related products have increased tremendously.
Perfluorinated compounds are commonly produced by electrochemical fluorination (ECF). As starting reaction for the production of PFOS related chemicals, the 3M corporation has introduced the fluorination of 1-octanesulfonyl fluoride to perfluorooctanesulfonyl fluoride (POSF), described in figure 1 (Anonymous, 1999).
Figure 1: Industrial formation of PFAS related chemicals via ECF (Anonymous, 1999).
Perfluorooctanesulfonyl fluoride (POSF) is thereafter used as starting chemical for a huge variety of different products within the PFAS group used in a vast array of products and applications. ECF produced PFAS related substances usually are containing a sulfonyl- or carboxylic group in their molecular structure.
During the past decades, consumers in western countries embraced these chemical products as important chemical tools for all types of household related processes (e.g., cooking, clothing, furniture, etc.) as described in a comprehensive report (Hekster et al. 2002). The chemicals of the PFAS group are characterized by carbon chains with variable lengths, to which fluorine atoms are covalently bonded. The strong C-F bond is yielding ultimately in virtually indestructible chemicals that until recently were thought to be completely biologically inert and, thus, not bioavailable for biochemical processes.
Since the late 1990s, increasing numbers of published scientific studies, pinpointing the potential environmental hazards posed by this type of chemicals, have brought PFAS related chemicals in the focus of international public environmental concern. Besides being an important industrial group of compounds, PFAS related residues are today considered as highly toxic, extraordinarily persistent chemicals that pervasively contaminate human blood and wildlife all over the world. In contrast to well-documented persistent organic pollutants like polychlorinated biphenyls (PCB), chlorinated-p-dibenzodioxins and furans (PCDD/F), governments and scientists are today especially concerned that the most pervasive and toxic members of the PFAS group will never degrade in the environment.
1-Octanesulfonyl fluoride
Perfluorooctanesulfonyl fluoride (POSF)
C8H17SO2F + 17 HF C8F17SO2F + 17 H2 4.5 – 7.0 V
ECF
13 of 107
Already in 2000, the U.S. Environmental Protection Agency (EPA) banned PFOS from the US market. Shortly thereafter, the manufacturer 3M also stopped voluntarily the production of PFOA. Currently, PFOA is also evaluated for regulatory actions by U.S.- EPA.
The production of sulfonyl-based fluorochemicals has recently been estimated around 6.5 million pounds world-wide only for the year 2000 (U.S. Environmental Protection Agency 2000). For PFOS only, a global production volume of 831 metric tons are estimated for the year 2000 (Anonymous 2002). Current environmental studies confirmed that PFAS are shown to be globally distributed, environmentally persistent and bioaccumulative (van de Vijer et al. 2003, Kannan et al. 2001 a,b, 2002 a,b, Giesy and Kannan 2001, Martin et al. 2004) with implication also for human exposure (Olsen et al. 2003, Taniyasu et al. 2003, Levitt & Liss 1986).
Under the joint leadership of UK and USA, the OECD co-operation on the Investigation of Existing Chemicals group performed a first risk assessment on PFOS and its salts in 2002 concluding that PFOS is persistent, bioaccumulative and toxic to mammalian species (Anonymous 2002). There are species differences in the elimination half-life of PFOS; the half-life is 100 days in rats, 200 days in monkeys, and years in humans. The toxicity profile of PFOS is similar among rats and monkeys. Repeated exposure results in hepatotoxicity and mortality (Anonymous 2002).
Also the Swedish Chemicals Inspectorate performed a preliminary national study on the risk evaluation status and possible national risk mitigation strategies in Sweden and other European countries on PFOS and its salts (Cederberg et al. 2004). The document reports on the strategy of the UK Department for Environment; Food and Rural Affairs (DEFRA) and the UK Environment Agency to develop a risk evaluation manual on PFOS and its potential to degrade under natural conditions (Brook et al. 2004). Based on this study, in UK the main application areas of PFOS and related compounds are as follows:
Impregnation/ waterproofing of textiles and fur products: 49 % (195 tons in 2001) Impregnation/ waterproofing of paper related products : 15 % (60 tons in 2001) Outdoor activities: 18% (70t in 2001) Flame retardants: 16% (65t in 2001) Others: 2.5% (10 tons in 2001) The European Union plan to adopt the resolutions according to the UK lead work on risk evaluation on PFOS and related compounds.
Also in the Netherlands a first evaluation of environmental risk posed by PFOS and related compounds revealed than 60 –105 tons/y (2002) are released through various products (Hekster et al. 2002).
Already in the 1980s, the acute toxicity of perfluorooctanoic acid (PFOA) and nonadecafluoro-n-decanoic acid (NDFDA) was evaluated in male Fischer rats (Olsen & Andersen 1983) and human B cells (Levitt & Liss 1986). In both studies toxic potency for PFOA and NDFDA was reported. Some of the toxic effects found for NDFDA were remarkably similar to those known for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). It was assumed that the acute toxicity of NDFDA might be due to the ability to interfere with fatty acid metabolism. Thus, further studies of its toxicity have been recommended as valuable in understanding toxicological mechanisms of action of TCDDs as well.
14
14
Due to the well-documented environmental concerns related to PFAS, the 3M Co-operation (St Paul, MN, USA) decided to discontinue its product line for most of the PFOS related products with the effective end of production occurring already around November 2001. In addition US-EPA has initiated regulation actions with respect to PFOS-related chemicals. However, the EPA’s initial actions and 3M’s phase out apply only to PFOS and its derivatives. Telomer-based PFAS (e.g. fluorotelomer alcohols) will continue to be produced and no disruption of supplies will or has been experienced so far. These telomer-based PFAS related compounds are chemical products of a telomerisation process where tetrafluoroethylene (TFE) as “building blocks” is used resulting in fluorinated alcohols with varying alkyl-chain length and usually one non-fluorinated carbon atom behind the hydroxylated final carbon atom.
US-EPA is currently assessing other perfluorinated chemicals like PFOA and related chemicals such as telomer products (Fluorotelomer alcohols: FTOH) with regard to possible environmental hazards and respective regulatory actions.
Currently, international research is concentrating on assessing the mechanisms of PFOS toxicity, identifying the full range of impacts in order to get a comprehensive understanding of contamination levels in the environment incl. people and wildlife.
Most of the selected PFAS compounds chosen for the presented screening exercise, are neither water nor lipid soluble, except for PFOSA, which is found to be surprisingly lipophilic and, thus, is expected to accumulate in the course of the food webs. Thus, relatively high levels were found in all biota samples.
2.1. Applications and product information A first survey of the product and application information available in the respective Nordic countries revealed that the excess of actual date is still difficult and often impossible for PFAS related compounds.
Norway is currently working on a material flow analysis. The preliminary estimates shows that fire fighting foams cover more than half the total PFOS-related substances and other PFAS used in the country. In protective coating and textiles the amounts is less than half of the total. In floorwax/polish and electronic industries the amounts are approximately 100-200 kg. The total amount for PFAS used in Norway is estimated with approximately 23 – 26 tons. However, there may be considerable uncertainties in these estimates (personal com., Ola Glesne). The total sales of PFOS based chemicals to the Swedish market in 1999 was 38 tons (3M, personal com.) most of which was used in textile and leather treatment and as industrial surfactant.
15 of 107
3. Samples for PFAS screening in the Nordic environment
Based on recently raised concerns about possible environmental impact of perfluorinated residues in pristine environments including Northern regions, the NTEM “Nordic Chemicals Group” of the Nordic Council of ministers has initiated a first screening on the presence and distribution of selected perfluorinated alkylated substances (PFAS) in the Nordic environment.
3.1. Sample selection: Criteria and priorities Based on already available scientific information on distribution and concentration levels, the NTEM “Nordic Chemistry group” in co-operation with the analytical laboratories in charge, the Norwegian Institute for Air Research (NILU, Kjeller, Norway) and the Institute for Applied Environmental Research (ITM, Stockholm, Sweden) selected a priority list of 6 compounds of the PFAS group (see table 1). In addition, the analytical laboratories have voluntarily added two priority compounds:
Perfluorobutane sulfonate (PFBS): This compound is announced as official successor for the PFOS related products by the manufacturers. According to the manufacturers, PFBS is not accumulating in biota. In accordance with this postulation, no PFBS traces were confirmed in biota samples yet.
Furthermore, perfluorodecane sulfonate (PFDS) has been included in the PFAS screening of biological samples.
A comprehensive selection of sample types spanning from seawater and precipitation towards marine mammals, as representatives for top predators, have been selected by NTEM, the working group of the Nordic council of Ministers.
Sampling locations are presented in two separate maps for sites were biota and abiotic samples were collected (see appendix 4).
For the abiotic environment, 7 sample types have been analysed and for the biotic environment, 16 species have been selected, representing freshwater and marine environments (table 2). Exclusively liver samples were analysed except for the fulmars from the Faroe Islands were eggs samples were pooled. The sample characteristics for the material provided by the 6 participating Nordic countries can be found in appendix 1.
16
16
Table 2: Sample types selected for the first Nordic PFAS screening.
The following marine and freshwater species were analysed for PFAS residues:
A total of 119 samples have been analysed covering a large and challenging variety of matrices. In addition, 13 abiotic and 3 biota samples have been collected as parallel samples in order to provide information about method robustness and variability (table 3).
Table 3: Parallel samples for method performance tests collected in all six
participating Nordic countries*
* Sample characteristics can be found in appendix 1
3.2. Sample collection As a guideline for appropriate sampling, the laboratories in charge provided a sampling manual for the sampling personnel in the Nordic countries participating in the screening exercise (Appendix 2). Detailed procedures for sampling, storage and transport were given. Sampling protocols for all sample types were implemented in the sampling manual.
Detailed information on the sampling locations are provided by the respective participating countries (see also appendix 1):
3.2.1. Finland
Abiotic samples: The waste water (sewage) treatment plants of Espoo, Helsinki and Porvoo are the biggest plants in Finland taking care of both industrial and municipal waste waters. The treatment systems are of the most modern quality and technique in Finland. In the plants there are treated the following amounts of waste water given in peq -values: Espoo = 235 000 person equivalent (peq), Helsinki = 815 000 peq , Porvoo = 170 000 peq. The landfill effluent studied was collected from the very large Waste Treatment Centre in Espoo, s.c. Ämmässuo landfill. The solid wastes are collected from the city areas of Helsinki, Espoo, Vantaa, Kauniainen and Kirkkonummi having the population over one million. The amount of waste received in 2002 was calculated to 670 000 tons. Biota: All Finnish pike samples were caught in the coastal water near the main urban city area of Helsinki (old City bay) about 200-500 m from the coastline. The area has
18
18
influences of polluted waters and sediment deposits since industrial PFA-substances were introduced into use.
3.2.2. Norway
Abiotic samples: The sampling of landfill effluents was performed on the vicinity of 5 landfill locations situated close to the Kristiansandfjord and the Oslofjord area. The landfill sites were selected in co-operation with the Norwegian State Pollution Control Authorities (SFT). Local-co-operation partners and scientists from the Norwegian Centre for Soil and Environmental Research (Jordforsk) suggested priority sites and the scientists of the institute prepared the sampling equipment. Samples were collected during the period July - October 2003. The rest of the abiotic samples were all collected in Lake Mjøsa or at sewage treatment plants with effluents to the lake (Appendix 1).
Biota: All freshwater fish samples were collected from the Norwegian Lake Mjøsa close to sewage treatment plants or potential source locations by scientists of the Norwegian Institute for Water Research (NIVA).
3.2.3. Sweden
Abiotic samples: Rain water samples were collected from the Råö research station. Sampling periods: RWSWE01: 02.10.2003 –11.11.2003, RWSWE02: 11.11.2003 – 03.12.2003., RWSWE03: 03.12.2003-15.12.2003. The freshwater sediment samples were collected from potential accumulation sited upstream and downstream the municipality of Kristianstad. Representative sewage sludge samples were collected over an entire week with representative daily sampling and unification after one week (week 39, 2003). Sewage was collected from the Sewage treatment plants (STP) Kistianstad Centrala ARV, Köpinge ARV and Tollarp (supposedly household waste water only).
Biota: Freshwater fish (Perch) was collected close to the freshwater sediment sites at the potential accumulation sitein the vicinity of the STP Kristianstad. Pooled samples containing in average 10 livers were provided. Atlantic cod were collected from one near shore site and one off-shore site (Hanöbukten, Hoburgen). Grey seal samples stem from three different locations from the Swedish Baltic coast (Gästrikland, Uppland, Sörmland)
3.2.4. Denmark
Abiotic samples: Aqueous and solid abiotic samples were collected in the vicinity of potential sources. Sewage sludge samples were chosen from large sewage treatment plants from the cities of Roskilde (Sewage treatment plant (STP) Bjergmarken), Odense (STP Stige) and Copenhagen (STP Lynetten). Seawater samples were collected in coastal areas close to the city of Rinkøbing. Sampling sites for sweater sampling close to the seage
19 of 107
effluent of Copenhagen (STP Lynetten), Odense (S1: Sedenstrand) were chosen. As reference site, the station 60 (st.60) situated relatively remote to effluent influences was also sampled.
Biota: All biota samples were collected in the vicinity of potential primary PFAS sources. Flounders were caught in the coastal water close to the capitol Copenhagen. Eelpout and herring were collected in the inner part of the Roskilde fjord. The harbour seal samples were collected close to settlements (Limfjord, Hesselø, Samsø and Øresund).
3.2.5. Iceland
Abiotic samples: Sediment and seawater samples were collected close to the landfill site Gufunes. Sewage sludge was sampled from the two STPs Klettagardar and Ananaust.
Biota: The Icelandic Minke whale samples were collected during a research expedition in the period August/ September 2003 in the North Atlantic waters (appendix 1). All marine fish samples were collected close to city of Reykjavik (capitol of Iceland). Long-rough dab, dab and sculpin samples were collected in the harbour areas of Reykjavik, close to potential PFAS sources (e.g.,Gufunes landfill).
3.2.6. Faroe Islands
Abiotic samples: Seawater, Sewage effluent, Sediment and Sewage sludge was collected from sites close to potential PFAS sources. Seawater samples were collected from the harbour area of Torshavn city (the capitol of the Faroe Islands). Two sediment samples were collected in the Torshavn harbour area and nearby Kaldbaksfjord. Sewage sludge was sampled from the STP Sersjantvikin in Torshavn.
Biota: Pilot whales were caught near shore at Sandagerdi in September 2002. The Arctic char samples stem from a landlocked population in a freshwater lake (á Mýranar). The fulmar eggs were collected from a bird cliff at Vidareidi. The Atlantic cod, sculpin and dab samples were collected in coastal waters close to settlements (Kaldbak, Kirkjubö and Torshavn)
3.3. Analysis and quantification All samples were sent by the national institution responsible for sampling directly to the analytical laboratories responsible according to the procedure described in the sampling manual (Appendix 2). The abiotic samples (sea water, precipitation, sediment, sewage sludge, landfill effluent, sewage effluent and freshwater) were sent for analysis to the laboratories of the Norwegian Institute for Air Research (NILU, Kjeller, Norway), whereas the biological samples (16 species) were analysed by the Institute for Applied Environmental Research (ITM, Stockholm, Sweden).
20
20
3.3.1. Preparation of sediment and sludge samples
All samples were stored at 6 ºC prior to sample preparation. No water content in sediment and sludge samples was determined. According to the supplement to the sampling manual (Appendix 2), the sewage and sediment samples should be sent to the laboratory in pre-tried or desiccated conditions. However, not all samples have been sent as required. Therefore, the samples were partly dried by filtering using a water-jet vacuum pump prior to analyses to ensure comparable sample conditions. Thereafter, quantification was based on a wet weight basis including the remaining water in the solid samples. Sediment and sludge samples were extracted using accelerated solvent extraction (ASE, Dionex, Sunnyvale, CA).
A stainless steal (11 ml volume) ASE cell was used for extraction. About 2 g rinsed and pre-cleaned sea sand (rinsed with methanol, then 200 °C for 4 h) was filled in bottom of the cell. Thereafter, approx. 5 g sample was transferred into ASE cell. Finally, the ASE extraction cell was completely filled with sea sand. Prior to extraction, 50 µl internal standard (ISTD) was added according to the standard listed in table 4.
Table 4: Internal and recovery standards used for PFAS quantification in aqueous
samples, sediment and sludge
Internal quantification standard (ISTD)
Recovery standard (RSTD)
0.1 ng/µl 7H-perfluoroheptanoic acid (7H-PFHpA) in methanol
0.2 ng/µl 3,5-bis(trifluoromethyl)phenyl acetic acid in methanol
F
FF
FF
FF
FF
FH
F F O
O
CF3
F3C O
O
Accelerated solvent extraction (ASE) was performed three times with methanol (17 min/cycle, 150 °C, 2000 psi). The three extracts were combined in a Turbovap container (Zymark, Hopkinton, MA, USA) and concentrated with a Turbovap evaporator to approx. 500µL. Final concentration to 200µL took place under gentle nitrogen stream (N2, 5.0 quality, Hydrogas, Porsgrunn, Norway).
Subsequently, 50 µL of recovery standard (see table 4) and 250 µL 4mM ammonium acetate buffer (in deionised water) was added. The closed sample vials were treated for 10 minutes in an ultrasonic bath. In order to remove precipitation, the samples were filtrated using a Microcon YM-3 centrifugal filter device (Millipore Corp., Billerica, MA) at 14 000 rounds per minutes (rpm) for 45 min. finally, aliquots of the sample extract were transferred into autoinjector vials for quantification with high-performance liquid chromatography coupled to mass spectrometric detection.
3.3.2. Preparation of water samples
About 500 ml of the collected water samples were used for PFAS analysis. Approx. 0.5 vol.-% formic acid (HCOOH, 2.5 ml for 500 ml water) was added to prevent uncontrolled microbiological degradation processes. As a first sample preparation step,
21 of 107
the water samples were filtered (using a water-jet vacuum pump) through a glass fibre filter (GFF: 142 mm, PALL Life Sciences, New York, USA, cat. no. 61635). GFF and aqueous sample are, thereafter separately prepared for analysis.
The GFF was folded into an 11 mL ASE-extraction cell. 50 µl ISTD (table 4) was added and extraction is performed according to the ASE method described earlier for sediment and sludge samples.
For water analysis, 80 mg ammonium acetate (approx. 2mM) and 50 µl ISTD was added. Finally the sample bottle (PET or PP) containing the water samples was weighed prior to solid phase extraction (SPE) using an analytical balance.
For solid phase extraction (SPE), HLB Plus SPE cartridges (0.25 g, Waters co-operation) were used. Prior to extraction, the SPE system was rinsed with approx. 20 ml methanol. The cartridge was thereafter conditioned using about 10 ml Milli-Q water. The water samples were extracted through the SPE cartridge with approx. 2 drops/sec. After sample extraction, the empty bottle (still containing small amounts of residual water) was again weighed. The weight difference is used as sample amount for concentration calculation.
Prior to elution, the SPE cartridge was rinsed with 2 mL methanol/water (40/60). PFAS elution is performed with 8 mL methanol. The methanol extract was, thus, concentrated to 200 µl by a combination of Turbovap and nitrogen concentration (see method description for sludge and sediment sample preparation) and 200 µl of a 4 mM solution of ammonium acetate in water was added.
Both filter and aqueous sample extracts were finally spiked with 50 µl RSTD, filtered if necessary as described for sediment and sludge extracts (in case of precipitation) and transferred to autoinjector vials for subsequent HPLC/MS analysis and quantification
3.3.3. Preparation of biota samples
As appropriate target tissue in all fish and marine mammal samples, liver was chosen. All samples were collected according to the sampling manual provided (appendix 2). The collected samples were thereafter send to the analytical laboratory (ITM) for quantification. The tissue samples were thoroughly homogenized before preparation of aliquots for analysis. Liver samples in excess of a total of 40 g were homogenized using a Braun turbo mixer (Braun, Kronberg, Germany). This procedure was also used for the preparation of pooled samples from individual samples delivered to the laboratory. Aliquots were then further homogenized in high-purity lab water (one part sample, 5 parts MilliQ-water; MilliQ, Millipore Corp.) using an Ultra-Turrax homogenizer. From this homogenate aliquots of 1ml were taken to analysis and for replicates as well as for matrix spikes.
The extraction method was based on ion pairing as described firstly by Ylinen et al. (1985) and further elaborated by Hansen et al. (2001) An aliquot of 1 mL of the homogenate was transferred to a polypropylene (PP) tube, mixed with 2 mL of 0.25 M sodium carbonate buffer, and 1 mL of 0.5 M tetrabutylammonium solution (TBA, adjusted with NaOH to pH 10). The resulting mixture was vortex mixed for 20 seconds. 5 mL methyl tertiary butyl ether (MTBE) was added and the tube gently turned for 20 minutes. The MTBE was quantitatively transferred to a second PP-tube and another 5 mL MTBE portion added to the first tube and the procedure repeated.
22
22
The combined MTBE extract was then gently evaporated until dryness using dry nitrogen (5.0 quality) and exactly 500 µL methanol added. All extracts were finally filtered through a 0.46 µm PP-filter prior to subsequent LC/MS determination.
3.3.4. Quantification of abiota samples
An Agilent high performance liquid chromatography system (HPLC, 1100 series, Agilent, Palo Alto, CA) in combination with a time-of-flight high resolution mass spectrometer (LC-TOF: Waters-Micromass, Manchester, UK) was used for quantification.
For analysis, 20 µl sample extract in methanol/water (1:1, v:v, 2 mM ammonium acetate) were injected. An ACE C18 reversed phase HPLC column (ACT, Aberdeen, UK; 150 x 2.1 mm, 3 µm particle size) was used for PFAS separation in abiotic samples.
For HPLC separation the following parameters were chosen:
Flow rate of the mobile phase: 200 µl/min;
Buffer: 2 mM ammonium acetate (NH4OAC) in both methanol (A) and water (B).
Gradient of the mobile phase: 0-1 min 50 % A; 1-6 min linear increase to 85 % A, 6-12 min isocratic 85 % A, 12-13 min linear change to 99 % A, 13-20 min isocratic 99 % A. Equilibration time 9 min.
The compounds of interest are transferred into the TOF high-resolution mass spectrometer. PFAS were quantified with electrospray ionisation (ESI) in negative ion mode (cone voltage –35 V; cone gass 10 l/hr; desolv. gas 400 l/hr; nebulizer max). Full scan mass detection in the range of m/z 200 – 720.
The following masses (Table 5) were used for PFAS quantification in abiotic samples collected from Nordic environments (water, sediment and sludge):
Table 5: Quantification masses (m/z) for the determination of selected PFAS in abiotic samples. Mass tolerance was typically 0.1 mass units.
Please note, the selected analytes differ between biota and abiotic samples (e.g., PFHpA, PFDcS was not analysed in abiota whereas PFBS was only analysed in abiota).
Representative mass traces for LC-TOF and LC-MS-MS quantification in environmental samples (landfill effluent and grey seal liver) are presented in figure 2 (a and b).
23 of 107
A:
B:
Figure 2: A.) Typical LC/MS-TOF mass traces for a landfill effluent sample. B.) Typical LC/MS-MS chromatogram for a grey seal sample Due to the different physico-chemical properties of the target compounds, especially for ionic compounds (sulfonates and acids) in sewage sludge, the method developed for PFOSA in sewage sludge is still considered as semi-quantitative but evaluated as sufficient for the here presented screening.
3.3.5. Quantification of biota samples
24
24
Identification and quantification was done by retention time using authentic reference compounds and selected reaction monitoring (Figure 2b: SRM-MS/MS, Micromass Quattro II, Altrincham, U.K.) with argon as a reaction gas and monitoring the transitions listed in table 6. The MS/MS instrumental conditions were identical to those described by Giesy and Kannan (2001). The detection limit was calculated from a low level spiked and extracted sample and was considered to be equal to three times the area noise level in each transition in the region of the target compound.
Table 6. Monitored MS/MS transitions
The LC system consisted of a C18 - column (2.1 mm x 50mm x 5µm, Hypurity C18, ThermoHypersil, U.K.) with methanol/water as the eluent and 10mM ammonium acetate as modifier. The eluent was delivered at a flow of 0.2 ml/min by a Waters Alliance 2690 pump (Waters Corp., Milford, MA. U.S.A.) with a gradient program starting at 40% methanol increased to 95% at 5 min, kept for 15 min and then returned to 40% at 20 min. Total run time was 35 min., including time for conditioning of the column. All calculations were done using multilevel calibration curves of respective matrix extracts spiked with external standard mixtures and using corrections for recovery of the different analytes. Generally, at least two spike levels were included for each new matrix and/or batch. Method blanks consisting of extracted MilliQ-water (Millipore, Billerica, MA, USA) were included with each batch and one field blank consisting of triolein was also analysed.
3.4. Quality control and method comparison Adequate quality control measures and documentation were introduced covering the entire analytical chain: sampling procedures, storage, transport, sample preparation and, ultimately analysis and quantification. For an appropriate sampling procedure with the aim to minimise contamination risk and document possible artefacts during sampling and transport, sampling protocol was developed in close co-operation between the analytical laboratories and the screening project’s steerning project.
Figure 3: Sampling protocol sheet for the collection of sea-, river and lake water samples for the Nordic PFAS screening project.
The sampling protocol (Appendix 2, sampling manual) was developed with the overall aim to address two major needs:
1.) Guidance for the personnel, responsible for sampling to avoid contamination.
2.) Ensure a complete documentation of the sampling procedure, quality of the sample and environmental circumstances during the sampling period.
Sampling protocols for seven major sample types were developed and distributed to the participating institutions. A typical sampling sheet is presented in figure 3,
26
26
the complete selection can be found in appendix 2 (sampling manual incl. supplementary information).
3.4.2. Limit of detection/ limit of quantification
Limit of detection (LOD) and limit of quantification (LOQ) are considered as two priority parameters, describing the quality of a quantitative analytical method. According to IUPAC (McNaught and Wilkinson 1997, Thomsen et al. 2003), the LOD, expressed as the concentration, cL, or the quantity, qL, is derived from the smallest measure, xL, that can be detected with reasonable certainty for a given analytical procedure. The value of xL is given by the equation: xL = xbi − + ksbi
where xbi − is the mean of the blank measures, sbi is the standard deviation of the blank measures, and k is a numerical factor chosen according to the confidence level desired.
For abiotic and biotic samples k = 3 (3 x signal/noise) was chosen for the present screening study.
According to guidelines developed by the “US great Lakes monitoring programme”, the limit of quantification is defined as the lowest concentration of an analyte that produces a signal/response that is sufficiently greater than the signal/response of lab reagent blanks to enable reliable detection (and thus quantification) during routine laboratory operating conditions. The analyte response at the limit of quantification (LOQ) should be at least 5 times the response compared to the blank response (www.epa.gov/glnpo/monitoring/data_proj/ glenda/codes/r_lim_tp.pdf).
During the present screening exercise, for PFAS analysis in abiotic samples the LOQ was set 5 x compound level in the highest field blank for the respective sample type. For PFAS determination in biota, the LOQ was defined as 3 x compound level in the highest field blank for the respective sample type.
LOD and LOQ determination was performed in accordance to the guidelines given in the above described documents.
3.4.3. Laboratory and field blanks
Laboratory and field blank analysis was an essential part of the quality control program. For all sample types relatively low laboratory and field blank levels have been determined except for PFOA where elevated concentration levels were found indicating possible contamination during transport and sample preparation (table 7 and 8).
Table 7: Laboratory and field blank determination for aqueous samples
27 of 107
The recovery rates for the abiota blank samples (table 7 and 8) was calculated for the internal standard (ISTD: 7H-PFHpA) against the recovery standard (RSTD: 3,5-bis(trifluoromethyl)phenyl acetic acid =100%) assuming representative behaviour of the ISTD compared to the unknown PFAS target analytes (see table 14).
The method limit of quantification (LOQ) was set as 5 x highest field blank levels determined (table 8) or, if no blank levels were present, as 5 x background noise in the chromatogram.
For biota standard spiked extracted matrix experiments have been performed to determine the respective recovery rates for all target analytes.
The detection limit was calculated from a low level spiked and extracted sample and was considered to be equal to three times the area noise level in each transition in the region of the target compound. The concentrations of target analytes were in the following range: low level spiked = 3-8ng/g; high level spiked = 30-80 ng/g.
Table 8: Laboratory and field blank determination for sediment and sewage sludge samples [ng/g ww].
- = not analysed.
The Finnish participants sent one field blank sample of triolein (standardised lipid: C57H104O6; CAS-nr. 122-32-7) to ITM. No target analytes were detected in this sample. The field blank was analysed in parallel with the Finnish pike samples with high recovery (50-80%) for all target analytes.
Analytical method blanks (MilliQ water) have accompanied each series and instrument blanks (MeOH) were included with each instrument series. Limit of detection levels are calculated as instrument detection limit at 100% recovery (table 9). The method LOD describes the lowest detection range depending on sample matrix (see 3.4.2).
For the PFAS analysis in biota, the recovery rates were found to be highly matrix dependent. The recovery ranges are listed in Table 10. Recovery is corrected for by quantifying each sample using a spiked and extracted sample from each series and matrix type.
Table 9: Limit of detection (instrument and method) for the target PFAS compounds in biota [ng/g].
Repeated injection of selected biota samples revealed the low variation in the quantification method for PFAS quantification in biota using LC-MS/MS techniques (table 11).
Table 11: Repeated PFOS quantification for Swedish and Finnish biota samples Sample Day 1 Day 2
COSWE06 9.1 10.9
COSWE07 8.1 9.0
COSWE08 8.7 9.0
COSWE09 20 24.5
COSWE10 6.4 8.4
COSWE06 6.4 11.0
PIFIN01 594 551
PIFIN02 220 204
PIFIN03 227 211
PIFIN04 258 240
PIFIN05 284 263
PIFIN06 273 253
PIFIN07 530 492
3.4.4. Performance tests (parallel samples)
As a part of an accompanying quality control programme abiotic samples (seawater, precipitation (rainwater), landfill effluent, sewage effluent, sediment and sewage sludge samples) as well as biota (long-rough dab, Arctic char and perch) have been collected in parallel to document the representativeness of sampling methods as well as analytical method variability and robustness of sample preparation (table 3). From 6 countries 6 different sample types (16 samples) were collected in parallel (2- 5 parallel samples respectively). The complete dataset can be found in appendix 3 (concentration lists).
29 of 107
Depending on sample type, sample numbers and absolute concentration the relative deviation varied between 3 % and 22 % for the method used for abiotic compounds. The variability in the parallel sampling test documents that the method repeatability is sufficient for the present screening exercise. In addition, sample types and amounts showed to be sufficiently representative for the screening exercise.
3.4.5. Laboratory intercomparison
For a first method intercomparison and as an integrated part of the quality control program, a standard solution (in methanol) was prepared for parallel quantification of PFAS compounds using the respective trace-analytical methods developed by the two participating Nordic laboratories. The results are summarised in table 13. Table 12: Method intercomparison between the NILU and the ITM method used for
PFAS determination and quantification (see method description). Concentration of the standard solution in methanol = Theoretical values
*please note, PFHxS is present as contamination in the PFOS standard mixture. Therefore, the theoretical value for PFHxS is probably underestimated.
For both methods deviations from the expected concentration value up to 55% have been found (table 12) demonstrating the results of the analytical methods used in this studies are considered acceptable compared with well-established and fully evaluated monitoring methods. However, a more comprehensive method comparison is needed to fully evaluate the analytical methods. The overall results of this first method intercomparison showed a good quality of the analytical and quantification methods sufficient for the here performed screening exercise with regard to accuracy and comparability of the method.
Thus, for future quality control and assurance tests, sample preparation methods as well as matrix related method intercomparison will be considered as additional necessary step in direction of a fully evaluated and quality assured trace analytical method.
Table 13. Differences between the NILU and ITM PFAS quantification method.
30
30
As already outlined before, both methods are developed following different concepts with regard to sample preparation, analysis and quantification. The major differences are identified for the final LC/MS analysis and quantification (Table 13).
3.4.6. Data handling and comparison
Compound- specific comparison of PFAS related residues in the various sample types is performed using median values instead of arithmetic mean since concentration distribution is not assumed to be normal distributed.
Topic Laboratory
NILU ITM
Standard calibration Internal and external quantification standard calibration
External matrix spike standard for calibration
Recovery Internal recovery standard Spiked extracted matrix
Instrumentation LC-TOF (ESI-) LC-quadrupole MS (ESI-)
Identification/ quantification
High resolution single m/z quantification
SRM-MS/MS transition
31 of 107
4. Results
The here presented data sets represent the first attempt to evaluate the distribution of PFAS related compounds in the Nordic environment. Institutions from 6 Nordic nations participated in the here presented screening exercise. A comprehensive selection of 23 different sample types representing abiotic and biotic environments were selected for the study (table 2).
Table 14A: Nomenclature for the selected abiotic samples within the Nordic PFAS Screening.
Sample type Abbreviation Country Abbreviation Sample Number
Sediment SD Finland FIN 10-12
Sediment SD Faroe islands FAR 5-7
Sediment SD Norway NOR 12-14
Sediment SD Sweden SWE 4-6
Sediment SD Iceland ICE 2 (a-e)
Sewage sludge SS Finland FIN 13, 14 (a-c) – 16
Sewage sludge
SS Norway NOR 15-16
Sewage sludge SS Denmark DAN 4-6
Sewage sludge SS Iceland ICE 3 (a-b) – 4(a-b)
Sewage Sludge SS Sweden SWE 7-9
Sewage sludge SS Faroe islands FAR 8
Seawater SW Finland FIN 1-4
Seawater SW Denmark DAN 1-2(a-b)-3
Seawater SW Iceland ICE 1(a-d)
Seawater SW Faroe islands FAR 1-3
Rain water RW Finland FIN 5-6
Rain water RW Sweden SWE 1(a-b)-3
Lake water LW Norway NOR 1-4(a-b)
Sewage effluent SE Finland FIN 8-9(a-b)
Sewage effluent SE Norway NOR 10-11(a-b)
Sewage effluent SE Faroe islands FAR 4
Landfill effluent LE Finland FIN 7(a-c)
32
32
Landfill effluent LE Norway NOR 5(a-b)-9
Sampling, sample handling, transport, sample preparation and quantification were performed along a detailed quality control program including sampling manual and standardised laboratory procedures ensuring the best possible data quality for the analysed Nordic samples. The nomenclature, chosen for the samples selected within the here presented Nordic screening exercise for perfluorinated organic contaminants (PFAS) is listed in table 14A and B.
According to the above given table 14A, the sample number is composed by the abbreviations of sample type, countries name and the sample number. Thus, the 4th seawater sample from Finland is named SWFIN04. Please note parallel samples are distinguished with letters (a-e).
Table 14B: Nomenclature for the selected biotic samples within the Nordic PFAS screening.
Sample type Abbreviation Country Abbreviation Sample Number
Pike PI Finland FIN 1-8
Pike PI Norway NOR 3
Perch PE Sweden SWE 1-3(a-b, 4
Perch PE Norway NOR 4
Burbot BU Norway NOR 1
Atlantic cod CO Sweden SWE 5-12
Atlantic cod CO Faroe islands FAR 5
Grey seal GS Sweden SWE 14-16
Harbour seal HS Denmark DAN 7-11
Trout TR Norway NOR 2
Flounder FL Denmark DAN 1-4
Eelpout EP Denmark DAN 5
Herring HE Denmark DAN 6
Long-rough dab LD Iceland ICE 1(a-c)-2
Shorthorn sculpin SC Iceland ICE 3
Shorthorn sculpin SC Faroe Islands FAR 1-2
Dab DA Iceland ICE 4
Dab DA Faroe islands FAR 3-4
Minke whale MW Iceland ICE 5-9
Arctic char AC Faroe islands FAR 6(a-b)-7
Pilot whale PW Faroe islands FAR 8-11
Fulmar FU Faroe islands FAR 12-13
33 of 107
According to the above given table 14B, the sample number is composed by the abbreviations of sample type, countries name and the sample number. Thus, the 4th pike sample from Finland is named PIFIN04.
Blank samples are additionally abbreviated with “BL”. Thus, the first field blank for Swedish sewage sludge is named SSBLSWE01 (see table 8).
4.1. Levels and distribution A comprehensive sample set consisting of 119 environmental samples:
- 11 seawater samples from 4 Nordic countries - 5 rainwater samples from 2 Nordic countries - 6 landfill effluent samples from 2 Nordic countries - 5 sewage effluent samples from 3 countries - 15 sewage sludge samples from 6 Nordic countries - 13 sediment samples from 5 Nordic countries - 4 Lake water samples from 1 Nordic country - 23 marine fish samples from 5 Nordic countries - 18 freshwater fish samples from 4 Nordic countries - 17 marine mammal samples from 4 Nordic countries - 2 pooled seabird egg samples from 1 Nordic country.
represents the basis for the here presented first investigation on the fate and distribution of PFAS related residues in the Nordic environment (table 2).
Although a large number of samples is represented, the large variety of sample types collected in each country and the small number of individual samples representing the respective sample type restrict the assessment of general trends and spatial distribution. The differences between sampling locations and sample types within a Nordic country may be as large as between the respective countries. These variations are not considered in the here presented screening study. Thus, the results presented herein, should only be considered and discussed as indications and hints for further in-depth scientific studies.
4.1.1. Abiotic samples
The abiotic subset of samples consisted of 59 samples ranging from seawater to sewage sludge and sediments (table 2). For a better comparison the abiotic sample set is divided into two sub-groups: abiotic solid samples and aqueous samples.
Solid abiotic samples The highest levels in solid abiotic sample material for PFAS related residues were found for sewage sludge samples (SSSWE09, PFOS = 2644 pg/g ww). Considerable concentration differences were found for the various sample types (figure 4).
34
34
Figure 4: SUM PFAS concentration for solid abiotic samples: Sediment and sewage
sludge from Nordic countries.
SUM PFAS = PFOSA+ PFHxS+PFOS+PFHxA+PFOA+ PFNA.
In general, sewage sludge samples have higher PFAS content compared to sediment samples. However, indications for surprisingly high variability in sewage sludge are found. The lowest contaminated sewage sample was from Finland (SSFIN16: 150 pg/g ww) and the highest from Sweden (SSSWE09: 3793 pg/g ww). Sum concentrations for sewage sludge in Finland vary between 150 – 2521 pg/g ww, Sweden between, 168 – 3793 pg/g ww, Norway between: 1048 – 1654 pg/g ww, Iceland around ~234 pg/g ww and Faroe islands around 1677 pg/g ww. The Denmark samples are found in the range between 650 and 1500pg/g ww. Country specific sewage sludge release pattern cannot be excluded as possible reason for these differences. However, numerous factors influence the final concentrations in sewage sample (e.g., differences in the concentrations in sewage effluents may be influenced by the amount of clean water passing through the plant). However, also the distribution patterns (figure 5) for the analysed sewage sludge samples reflect the variability of the PFAS compounds found in this specific matrix.
35 of 107
Figure 5: Compound specific levels for PFAS residues in solid abiotic samples:
Sediment and sewage sludge. Concentration levels are given in pg/g ww (for more sampling details see appendices).
PFOS and PFOA are the predominant PFAS residues in solid abiotic samples. However, in the Finnish sewage sludge samples (see appendix 3: SSFIN13 and SSFIN14) also PFHxA is contributing significantly to the PFAS burden. No detectable PFAS residues were found in the Swedish freshwater sediment samples except for one sample were PFOS was found (SDSWE04: PFOS = 69 ng/g ww). In the only Icelandic sediment sample all PFAS were below limit of quantification (LOQ).
Aqueous samples For the here presented screening study five different types of aqueous sample types were chosen: Seawater, lake water, rain water, sewage effluent and landfill effluent (table 2). Remarkable concentration differences were found for the different sample types (Figure 5). Highest levels were found for Norwegian landfill effluent samples (LFNOR06 ; SUM PFAS = 1537 ng/L, LFNOR08; SUM PFAS = 1162 ng/L).The Finnish landfill effluent samples were collected in parallel and reflect again the good repeatability of the analytical method (see chapter about quality control). Sewage effluent had considerable lower concentration levels (SUM PFAS, SENOR11 = 59 ng/L and SEFIN08 = 105 ng/L). Lowest concentrations were found for lake water and seawater samples (Figure 5) and rainwater levels ranging between sewage effluent and lake water concentrations. In all aqueous samples regardless sample type PFOA represented the predominant PFAS compound (expect for the highest contaminated Landfill effluent sample LFNOR06 where PFHxA was the predominant PFAS related residue). PFOS, the most prominent PFAS residue in solid abiotic samples, is not as
36
36
dominant in the aqueous samples. However in Finnish landfill effluent samples PFOS is found in the same concentration range as PFHxA.
Figure 6: SUM PFAS concentration for aqueous abiotic samples: Seawater, rain water,
lake water, landfill effluents and sewage effluents. Concentrations are given in SUM PFAS = ng/L. SUM PFAS = PFOSA+PFBS+PFHxS+PFOS+PFHxA+PFOA+ PFNA.
For more information on the sample characteristics see appendices 1 and 3.
4.1.2. Biota
For comparison, the biota samples collected for the Nordic PFAS screening are presented as four sub-groups: Freshwater fish, marine fish, seabirds and marine mammals. A total of 60 biota samples are collected whereof 23 are marine fish samples, 18 freshwater fish, 17 marine mammals and 2 marine bird samples (table 2). The comprehensive selection of biota samples represents various biological and environmental endpoints. Thus, differences in levels and pattern distribution are expected to be related to trophic level, food habit, habitat, age, sex etc (biological information as well as other sample characteristics are summarised in appendix 1). Freshwater fish In all freshwater fish (liver), PFOS is the predominant PFAS representative (figure 7) followed by PFOSA regardless trophic level except for one sample (PINOR03; PFOS = 24 ng/g ww, PFOSA = 60 ng/g ww) where PFOSA is the most prominent compound. Pike represents the freshwater top predators and, thus, shows the highest PFAS contamination in the analysed freshwater biota. The highest PFAS levels were found in Finnish Pike samples (PIFIN01, PFOS = 551 ng/g ww, PFOSA = 141 ng/g ww). PFOS and PFOSA represent in all samples more than 90% of the PFAS burden. Although the Swedish perch samples represent a lower trophic level as pike, the PFOS levels are not significantly lower than found for Pike. However, the PFOSA
37 of 107
contribution is significantly minor compared to pike pointing towards food chain specific uptake or differences in transformation processes (Figure 7).
Figure 7: Compound specific level distribution for PFAS residues in freshwater fish
liver: Pike (Fin=8, NO =1), perch (SWE=4, NO=1,), trout (No=1), landlocked Arctic char (FAR = 2) and burbot (NO=1). Concentrations are given in ng/g ww. For more information on the sample characteristics see appendices.
Norwegian and Swedish perch showed comparable PFAS concentration and patterns. However the contamination level in the one Norwegian pike sample is considerably lower as found for the Finnish pikes. In addition, PFOS is not the dominating PFAS residue (but PFOSA) in the Norwegian Pike sample as already mentioned before. In addition, the fact that PFOS and PFOSA were also detected in anadromous Arctic char in the Faroe Islands indicate that long-range transport in air and/or precipitation is occurring. Marine fish The PFAS distribution in marine fish species (liver samples) is characterised by a surprisingly high variability reflecting differences in food chain, food habits and transformation strategies. Also for marine fish species, PFOS represents in most cases the predominant representative of the PFAS contamination (figure 8). However, in
38
38
Sculpins, PFOSA is the found as the highest PFAS contaminant. Also for the Atlantic cod sample from the Faroe islands PFOSA is higher concentrated than found for the Swedish cod samples (Figure 8). The highest PFAS levels were found for a Swedish cod (COSWE12; PFOS = 62 ng/g ww) and a Danish Eelpout (EPDAN05; PFOS = 60 ng/g ww). Danish Eelpout and Herring contained also measurable amounts of minor PFAS contributors like PFHxA and PFOA. In Long-rough dab and sculpin from Iceland PFHxA is even one of the dominating PFAS contributors. However, PFHxA is not prevalent in the Icelandic dab sample. The marine fish samples from the Faroe Islands are the lowest contaminated samples measured with regard to PFAS. Country specific application pattern can, thus, not be excluded as important reason for pattern distribution and concentration levels. .
Figure 8 Compound specific level distribution for PFAS residues in marine fish liver
[ng/ ww]. PFAS was determined in Atlantic cod (SWE=8, FAR=1), Herring (DAN=1), Flounders (DAN = 4), Eelpout (DAN =1), Long-rough dab (ICE = 2), Sculpin (ICE=1, FAR=2), Dab (ICE=1, FAR =2).
Nevertheless, marine fish species are significantly minor contaminated compared with the previously described freshwater samples indicating a certain correlation to the primary source distance. Please not that Perfluorodecane sulfone (PFDS) levels are surprisingly high in all Icelandic marine fish samples.
Marine mammals The 17 marine mammal samples (liver samples) selected for the here presented PFAS screening represents the top predators of the marine environment. Marine mammals are considerably higher contaminated as the marine and fresh water fish samples previously described. However, Icelandic minke whales are characterised by a relatively low PFAS contamination compared with the pilot whales from the Faroe Islands (figure 9) indicating a correlation to the position in the food chain and food habits. Swedish and Danish grey seals are highest contaminated with PFAS residues. The highest PFOS values were found in Danish Harbour seal samples (HSDAN10;
39 of 107
PFOS = 977 ng/g ww). Usually PFOS is the dominating PFAS residue also in marine mammals. However in Pilot whales from the Faroe Islands, PFOSA is equally contributing to the PFAS burden in two cases even exceeding the PFOS levels considerably (PWFAR08; PFOSA = 218 ng/g ww and PWFAR09; PFOSA = 364 ng/g ww).
Figure 9: Compound specific level distribution for PFAS residues in marine mammals
(livers): Harbour seals (Denmark), grey seals (Sweden), minke whales (Iceland) and pilot whales (Faroe Islands).
In all marine mammal samples, PFDS, a perfluorinated residue with a long-alkylated chain, is found in considerable amounts indicating bioaccumulation potential for this type of PFAS contaminants. Also hexanoic compounds (PFHxS and PFHxA) are found in measurable concentrations in all sample types (highest in harbour seals).
Marine birds Two Fulmar samples (pooled eggs) were collected as pooled samples at the Faroe Islands. The PFAS levels were found in the sample concentration range as for marine fish species (Figure 10). Although PFOS is also for Fulmars the predominant PFAS residue, PFOSA is not as dominant as found for the other biota samples. In Fulmars, PFNA is the second highest PFAS contaminant. Nevertheless, PFOS stands for about 95 % of the PFAS contamination in the Fulmar samples from the Faroe
40
40
Islands.
Figure 10: Compound specific level distribution for PFAS residues in pooled Fulmar
egg samples from the Faroe Islands [ng/g ww].
41 of 107
5. Discussion and recommendations
The first Nordic screening on fate and distribution of PFAS related residues in environmental samples confirmed that PFAS residues are widely present in the Nordic environment. The first overview indicates that sewage sludge and landfill effluents are contributing as important anthropogenic sources for the release of PFAS into the environment. Large amounts of PFAS related contaminants are, thereafter, deposited in sediment. Whether remobilisation from the sediments into the adjacent water column and the food web also is an important contribution remains still unknown at present. However, significant amounts of PFAS related chemicals mostly dominated by PFOS are accumulating in the marine and freshwater food web into the top predating organisms reaching surprisingly high concentration levels.
5.1. Source elucidation Already during preparation of the presented Nordic PFAS screening, sewage and landfill was identified as major anthropogenic sources based on general literature survey. Thus, sewage sludge, sewage effluent and landfill effluent was collected to document the concentration burden in this specific sample types representing primary environmental releases (figure 11-13).
Figure 11a: Median distribution [pg/g ww] compounds in Nordic sewage sludge
samples.
42
42
Figure 11b: Percentage distribution of PFAS related residues in Nordic abiotic solid
samples (SS = sewage sludge and SD = sediment).
A first comparison of PFAS contamination in sewage sludge collected in 6 different Nordic countries revealed significant pattern differences. In general PFOS was the predominant PFAS residue in sewage sludge (figure 11a-b). On the other hand, PFOA, is the predominant PFAS residue in some Finish, Danish and Faroe Island sewage sludge samples (SSFIN15, SSDAN06, SSFAR08). In few samples (SSFIN13, SDNOR12, SDFAR07) even PFHxA is dominating the PFAS pattern. These strong local variations in the PFAS distribution and levels indicate different sources or considerable temporal fluctuation in the release patterns.
Highest individual PFAS concentrations were found for a Swedish sludge sample (SUM PFAS SSSWE09 = 3793 pg/g ww, figure 4). The detected differences in pattern distribution point towards an application specific release pattern from anthropogenic sewage sludge sources into the environment with country specific influences. The PFAS distribution in sewage effluent samples reflect in general the concentrations found in sewage sludge from the different Nordic countries (Figure 12).
However, the median concentrations in the three sewage effluent sample sets are surprisingly homogeneous distributed (Figure 12). In both Norwegian and Finnish effluent samples, PFOA is the predominant PFAS constituent whereas PFHxA is equally concentrated as PFOS in the Finnish sewage effluent sample The sewage effluent sample from the Faroe island is considerably lower contaminated. The obvious shift in PFAS predominance from sewage sludge to sewage effluent (especially pronounced for the samples from the Faroe islands) indicated that also the physico-chemical properties are important in the distribution and release of PFAS from primary sources like sewage sludge. PFOA and PFHxA are more soluble in aqueous samples than PFOS. Therefore, PFOA and PFHxA are distributed via the water phase whereas PFOS is assumed to be introduced into the environment mainly via biosolids adsorbed on surfaces. PFBS, the perfluorinated chemical promoted currently as successor for PFOS related products is found in significant amounts in all aqueous samples.
43 of 107
Figure 12: Median distribution [ng/L] compounds in sewage effluent
Landfill effluents are identified as additional possible anthropogenic sources (figure 13).
Figure 13: Median PFAS distribution [ng/L] in Nordic landfill effluent
The concentration levels found for the two sample sets of landfill effluent are considerably higher than the levels previously discussed for sewage effluents (around 10 x higher, see figure 12) confirming the significant contribution of landfill effluent water as major anthropogenic source for PFAS in Nordic countries. PFHxS, PFHxA and PFOA are dominating the Finnish landfill effluent sample. In the Norwegian samples, PFOS, PFHxA and PFOA are dominating. PFOA is the predominant PFAS constituent in all landfill effluent.
44
44
5.2. Sample specific patterns and fate estimation The distribution of PFAS related residues in environmental samples is ruled by a combination of chemical, physical and biological processes including physico-chemical properties of the chemical, the physical environment (temperature, adsorption, storage medium, light-conditions) and biological accumulation and transformation processes.
5.2.1. Potential sources
Sewage sludge, sewage effluent and landfill effluents are identified as potential primary sources for the release of PFAS residues into the Nordic environment due to the high levels of PFAS related compounds detected in these sample types (figure 5 and 6).
Figure 14: Relative pattern distribution of PFAS residues in Nordic sewage effluent
and landfill effluent samples.
Based on the available data, the release pattern seems mainly governed by physico-chemical properties of the target compounds in combination with application patterns characteristic for the respective Nordic country.
The PFAS patterns in sewage sludge (figure 11b), sewage effluent and landfill effluents are mainly ruled by PFOA, PFOS and PFHxA (figure 14). Sewage sludge from Finland is dominated by PFOA whereas Swedish sewage is dominated by PFOS. Norwegian sewage effluent samples are dominated by PFOA and PFOS. Sewage effluents from Finland do not reflect the characteristic patterns found for the respective sewage sludge samples indicating a strong “day-to-day” fluctuation in the release patterns and amounts.
Relative high concentration levels were found in marine mammals, which usually express strong migration behaviour. In addition, the concentration ranges reported on a global basis (Giesy et al. 2001) in combination with the here presented data set indicate that already now many PFAS related residues (PFOS, PFOSA) must be considered as ubiquitously distributed throughout the globe and thus, long-range transport (e.g., ocean currents, atmosphere) is expected to be a major source for PFAS contamination in remote places such as the Arctic regions.
45 of 107
5.2.2. Seawater, fresh water and rain water samples
Adjacent water bodies serve as direct recipients of the sewage and landfill related environmental release. However, except for one seawater sample (SWFIN01), the water samples (seawater, lake water, precipitation) do not reflect the primary release pattern identified for the previously described sewage and landfill samples (figure 14 and figure 15). The PFAS pattern in all Swedish and Finnish precipitation samples (rainwater) are governed by PFOA (Figure 15).
Figure 15: A: Median concentrations for seawater; B: rain (Sweden) and lake water/
fresh water samples (Finland, Norway). Concentration are given in ng/L.
Whereas no clear pattern distinction between lake water, rain water and seawater could be established regardless location (where the samples have been collected).
5.2.3. Sediment
The median distribution (Figure 16) showed that the PFAS levels in sediment are different in the respective Nordic countries. PFOS is only dominating in the Finnish
A.:
B
46
46
sediment samples. In the Norwegian samples, PFOA is the predominant PFAS residue, whereas the lowest levels were found in sediments from Sweden with PFOS as the predominant PFAS constituent. Low PFAS levels were also found in Faroe Island sediments with PFOSand PFHxA as dominating compound.
PFOS was identified as important PFAS related pollutant in sediments representing in all samples the predominant or the second most concentrated PFAS contaminant in all sediments analysed.
Whether these distinct differences are a result of the differences in sediment composition or a result of country specific release patterns remains unknown und should be investigated in more detail.
Figure 16: Median PFAS concentrations in Nordic sediment samples.
5.2.4. Biota
A complex combination of trophic level, up-take via the food web, passive accumulation processes as well as bioconcentration is assumed to rule the PFAS patterns in the Nordic biota samples.
47 of 107
PFAS pattern in the various biota samples reflect the biological status of the species and not so much the sample location/ country. A first pattern analysis revealed that species representing high trophic levels (top predators) are mainly dominated by
Figure 17: Reative pattern distribution for PFAS residues in freshwater fish (A), marine fish (B) and marine mammals (C). For samples codes see page 33-34
A:
B:
C:
48
48
PFOSA and PFOS (figure 17a-c). Danish harbour seals and Swedish grey seals are dominated by PFOS whereas PFOSA is predominant in Pilote whales from the Faroe Islands. Also Finnish pike samples are strongly influenced by PFOS. Marine and freshwater fish species representing lower trophic levels seem not as strongly influenced by PFOS and PFOSA as top predating species.
PFOS and PFOSA are present in the fresh water pike and perch in almost as high concentrations as in the marine mammals. These high burdens are probably due to primary source in the vicinity of the sampling site. However, the diversity of PFAS found in the marine fish and mammals seems to be higher than in the fresh water fish reflecting a stronger influence of possible secondary sources like long-range transport (atmosphere and ocean currents) as well as migration patterns of marine species (predators and prey species).
5.3. Spatial distribution A first attempt to find indications for country specific distribution pattern is undertaken. Only for sample types collected and analysed from more than 3 Nordic countries a comparison of the results has been performed. However, statistically the available data set is by far not sufficient to reveal significant and statistically confirmed spatial trends. Thus, the here presented tendencies and patterns should only be considered as indications and used as basis for later in-depth trend studies.
5.3.1. Seawater
Figure 18: Spatial distribution of PFAS residues (ng/L median concentrations) in seawater from Nordic countries.
Please note, the placement of the histograms in the map does not reflect any information on sampling location.
In general, the PFAS median values for seawater samples collected in Iceland, Faroe Islands, Denmark and Finland reflect a similar PFAS distribution pattern (Figure 18).
ng/L
49 of 107
PFOA is dominating all seawater samples analysed (up to 70% of the PFAS burden) usually followed by PFHxA. However, in the Finnish samples, PFOS is slightly higher concentrated than PFHxA. The overall concentrations for PFAS are highest in Finland and Denmark indicating a certain influence of the population density. However, The concentration differences between the countries with smaller populations (Iceland, Faroe Islands) and larger populations (Denmark, Finland) are surprisingly small (around factor 2). Thus, the contribution of PFAS burdens from long-range transport may be considered and investigated as import source in future investigations.
However, sampling specific circumstances (e.g., distance to potential sources, sea currents, tidal conditions etc.) may contribute to the small differences between PFAS levels found in seawater.
5.3.2 Sewage sludge
For many comprehensively investigated anthropogenic pollutants, sewage sludge is considered as a significant primary release source. Therefore, also for the here presented study, sewage sludge was considered as an important sample type and, thus, collected in all six participating countries. A direct correlation between population size contributing to the sewage treatment plant (person equivalents) or specific sewage treatment procedures with the PFAS pattern found in sewage samples was not established in the here presented study. However, indications reflecting possible country specific application and release patterns are obvious (figure 19). The concentrations found in sewage sludge are relatively high amounting up to the lower micro-gram per gram (wet weight) range.
Figure 19: Spatial distribution of PFAS residues (pg/g ww median concentrations) in
sewage sludge from Nordic countries.
pg/g ww
50
50
Please note, the placement of the histograms in the map does not reflect any information on sampling location In the median distribution presented in figure 20, usually PFOA and PFOS were the predominant PFAS residues released through sewage sludge in all Nordic countries. However, whereas PFOA dominated in sewage samples from less populated countries (Iceland, Faroe Islands), PFOS was more prevalent in sewage from Denmark, Norway and Sweden). The PFAS burden of sewage from Finland, on the other hand, showed a somewhat different pattern. PFHxA is identified as second high concentrated PFAS residue in those samples indicating different release patterns (Figure 19). PFNA was also found insignificant amounts in Norwegian, Danish and Finnish Sewage (up to 10% of the PFAS burden).
When comparing concentration ranges for PFOS, relatively similar concentration ranges were confirmed for Nordic countries with large population numbers (Norway, Sweden, Denmark and Finland). For Iceland and the Faroe Islands very low levels were found (table 15).
Table 15: Concentration ranges for sum PFOS in Nordic sewage sludge samples
Country Number of sampling locations Concentrations [pg/g ww]
Norway 2 449 - 1023
Denmark 3 316 -1041
Sweden 3 167 – 2644
Finland 3 35 - 925
Iceland 2 < LOQ - 220
Faroe Islands 1 241
5.3.3. Sediment
Sediment samples were collected from five Nordic countries. In the Icelandic sample no PFAS were detected and only in one Swedish sample PFOS was at a quantifiable level. PFAS levels found in Norwegian Freshwater sediments were surprisingly high with up to 637 pg/g ww (sum PFAS median concentrations) for these sediment samples (figure 20).
However, a spatial comparison of PFAS levels in sediment samples is difficult since in general sediment composition, sedimentation rate, depth of the water column, etc. play as significant role in the up-take rates and residence time of persistent contaminants in those environmental compartments.
In sediments from the Faroe Islands, Sweden and Finland, PFOS is the predominant PFAS residue followed by PFHxA. In Swedish samples, even PFOS only was detected. The Norwegian and Swedish sediment samples were collected from freshwater lakes. The Norwegian sediments were collected from the largest freshwater lake in Norway, lake Mjøsa. The Swedish samples showed exclusively PFOS as dominating residue. The Norwegian sediments, however, are dominated by PFOA, with considerable contributions of PFOS and PFHxA. Also PFNA was founding at significant levels. It can be assumed that, hydrological conditions (tidal patterns, currents), sediment composition, sedimentation rates etc. are significantly different in freshwater sediments compared to the ocean-borne sediments from the Baltic Sea
51 of 107
(Finland), and the Atlantic Ocean (Faroe Islands and Iceland). Thus, no ocean current long-range transport contributes to and no tidal currents dilutes the PFAS burden deposited in the fresh water sediments from Lake Mjøsa (Norway) and the Swedish sites, whereas ocean-borne transport can be assumed contributing significantly to the PFAS patterns and burden in sediments from Finland and the Faroe Islands.
Figure 20 Spatial distribution of PFAS residues in sediment (pg/g ww median
concentrations) from Nordic countries.
*) In the Icelandic sediment samples analysed, very low amounts of PFAS were found (see appendix 3).
Please note, the placement of the histograms in the map does not reflect any information on sampling location
5.3.4. Biota
Specimens of 16 different species were collected representing marine and fresh water environments. Samples from 6 species were collected from 2 different Nordic countries. Therefore, a comprehensive spatial distribution analysis is not possible based on the biota samples collected for the here presented Nordic PFAS screening exercise. However, the restricted data available are useful as indications for level distribution of PFAS in biota which should be subject for further elucidation.
In pike collected from Norway and Finland, the total PFAS concentration and contribution of PFOS to the overall PFAS burden is different (figure 21) indicating differences in primary sources and/or feeding habits.
pg/g ww
*)
52
52
Figure 21: Median distribution of PFAS in Pike liver [ng/g ww] from Finland (n=8) and Norway (n=1).
The Finnish pike samples are higher contaminated with PFAS as the Norwegian sample. However, whereas the PFOSA levels are in a similar concentration range in pike liver from both countries, PFOS contributes with 10x higher levels to the overall PFAS burden in the Finnish pikes. The concentration levels in these freshwater fish were surprisingly high, found in the same concentration range as for top predating marine mammals (grey seal liver), indicating a considerable release of PFAS into freshwater system pointing towards different contamination sources. However, these preliminary assumptions are based on a very small data set and, thus, should only be considered as indication.
The levels in Atlantic cod were found in general to be low (low ng/g ww range). The Swedish cod samples were collected in the Baltic Sea, a region where many potential primary source are expected to release PFAS related residues in the marine environment close to the natural habitats of the cod. Thus, the levels of PFAS (median values from eight samples) in the Swedish cods were about 2 times higher than in the cod liver sample from Torshavn harbour (Faroe Islands, Figure 22). Whereas PFOS was the predominant PFAS contributor to the Swedish samples, PFOSA dominated in the Cod sample from the Faroe Island indicating different sources for respective populations.
53 of 107
Figure 22: Median distribution of PFAS in Atlantic cod liver [ng/g ww] from the Faroe
Islands (n=1) and Sweden (n=8).
In all other biological samples analysed for PFAS residues, no clear signals for county or region specific PFAS patterns and levels were identified.
5.4. Comparison with literature data Until today only few reports are published and, thus, publicly available on the distribution of PFAS related residues in environmental sample. A first attempt is, therefore, made to compare the results of the Nordic PFAS screening with already available literature data reporting PFAS levels in comparable environmental matrices. However, due to the comprehensive analytical challenges posed by this specific group of environmental residues, the trace analytical methods available today are still in the stage of development. Thus instrumental differences in analytical methods used, method validation documented and quantification methods applied restrict a comprehensive comparison of the literature data available. Therefore, the here presented comparison should only be considered as indicative. More up-to-data international investigations (like the here presented Nordic screening exercise) are on the way, which, hopefully, will allow in future a more thorough comparison and evaluation of global distribution, fate and occurrence of PFAS related environmental contaminants.
5.4.1. Water
Several published scientific papers report on PFAS reported contaminants in aqueous samples. Three international studies from Germany, USA and Japan are identified which reported environmental concentration of PFAS residues (table 16).
54
54
Table 16: Concentration distribution of selected PFAS residues in aqueous samples [ng/L]
Matrix Location PFHxS PFOS PFOA PFOSA Reference
Surface water (freshwater)
Germany - 2 – 43 LOD – 8 - Lang et al. 2004
Tennessee upstream of fluorofactory
USA, Tennessee river
- 17 – 54 <25 - Hansen et al. 2002
Marine surface water
Japan <11 LOD – 59 - - Taniyasu et al. 2003
Nordic sea water
Nordic countries
0.08 – 4.39 <LOD – 21.7 3.53 –8.48 <LOD - 0.07 This study
Nordic lake water
Norway <LOD – 0.11 <LOD – 0.48 4.82 - 8.23 <LOD This study
Nordic precipitation
Nordic countries
<LOD – 0.59 0.24 – 2.97 8.23 – 16.8 <LOD - 0.14 This study
The German and the US study reported PFAS levels in freshwater, whereas the Japanese study focused on PFAS residues in marine surface water samples. In the literature data on fresh water used for this comparison PFOS was the predominant PFAS constituent. No distinct differences between freshwater and seawater (surface) is found in the literature references. These results are different compared to the here presented Nordic data set. However, the literature values focus mainly on PFOS and no information about e.g. PFHxA is available by now. In all Nordic water samples (seawater, freshwater (lake) and precipitation), PFOA is the predominant PFAS residue followed by PFOS or PFHxA. The PFOS levels in the Nordic samples are considerably lower than reported in the three reference studies whereas PFOA values were in comparable concentration ranges as described in the literature (table 15). However, the relative high LOD for PFOA reported in the USA study (Tennessee) indicate considerable methodological limitations.
5.4.2. Sediment
Only one recent study has been identified in the literature so far reporting on PFOS and PFOA residues in sediment samples (table 16). A first comparison of these literature data with the here reported Nordic sediment data revealed that the sediment concentrations measured in the Nordic samples were considerably higher compared with the sediment samples analysed from The Netherlands. In-line with the literature data from the Netherlands PFOS is identified as dominating PFAS residue in sediment samples (table 17). However, a direct comparison of the two data sets is difficult because the Dutch sediment concentration levels are reported on dry weight basis, whereas the Nordic data set is based upon wet weight. In general, the Nordic sediment samples were higher contaminated than reported for the Dutch sediments, revealing considerable contamination potential in Nordic sediments.
55 of 107
Table 17: Concentration distribution of selected PFAS residues in sediment samples [ng/g]. dw = dry weight, ww = wet weight Matrix Location PFHxS PFOS PFOA PFOSA Reference
Sediments/ suspended matter (dw)
The Netherlands
- LOD – 47 LOD – 24 - Schrap et al. (2004)
Nordic Sediment (ww)
Nordic countries
LOD – 45 LOD – 892 LOD – 312 < LOD This study
5.4.3. Sewage sludge
Also for PFAS in sewage sludge only one relevant study (Germany) was identified for comparison with the here presented Nordic PFAS study. However, PFOS was analysed in the German study and the detection limit was extremely high (6000 ng/g dw). Thus, PFOS was not detected in any of the German samples. In the Nordic study, PFOS was the dominant PFAS related contaminant in sewage sludge, but the highest concentration found was below the detection limit of the German study, which, furthermore, is given on dry weight basis. This prevents any comparison of the two studies. In the Nordic sewage sludge samples PFOA was detected as the second dominant PFAS residue, whereas PFOSA and PFHxS were detected in levels up to 90 ng/g ww (table 18).
Table 18: Concentration distribution of selected PFAS residues in sewage sludge samples [ng/g]. dw = dry weight, ww = wet weight Matrix Location PFHxS PFOS PFOA PFOSA Referene
Sewage sludge (dw)
Germany - <LOD (6000 ng/g dw)
- - Schröder (2003)
Nordic sewage sludge (ww)
Nordic countries
<LOD – 91 55 – 2644 <LOD – 1075
<LOD – 94 This study
5.4.4. Biota
Most of the PFAS data found in scientific literature were reported for various biological samples. Based on this so far reported literature survey, an ubiquitous global distribution must be assumed for PFOS and related compounds (table 19). A set of 9 relevant scientific references reporting PFAS residues in biota was identified for the here presented literature comparison. One common feature was identified for all reported biota data. PFOS was the predominant PFAS residue identified in biota regardless sample type or trophic level. A first comparison confirmed, that the highest environmental PFAS burdens were found in marine mammal samples (table 19), although highest individual PFOS values were found for single marine fish samples. Taniyasu et al. (2003) reported up to 7900 ng/g ww in a Japanese fish liver.
Also for marine mammals, mainly PFOS was reported in international studies except for a study on Baltic seals were also residues for other PFAS related compounds were found (table 19). The PFAS (e.g., PFOS, PFHxA, PFOA, PFOSA) concentrations found in the Nordic seal samples (Denmark, Sweden) are found in the same concentration range as reported from the Baltic study. In the Nordic study, all seal samples have been found to be highest contaminated with PFAS residues of all Nordic biota analysed.
56
56
Table 19: Concentration distribution of selected PFAS residues in biological samples [ng/g]. dw = dry weight, ww = wet weight.
Fish liver (ww) worldwide - LOD – 170 - - Giesy et al. (2001)
Fish liver (ww) The Netherlands, Western Scheldt
- >10 – 7760 - - Hoff et al. (2003)
Oysters (dw) USA - LOD – 1225
- - Kannan et al. (2002d)
Eel filet (ww) The Netherlands - LOD – 143 <LOD - Schrap et al. (2004)
Atlantic salmon liver (ww)
Not identified <7.5 <8 <19 <19 Kannan et al. (2002a)
Fish liver (ww) Japan LOD – 19 3 – 7900 - - Taniyasu et al. (2003)
Freshwater fish (ww)
Nordic countries
<LOD – 3.4 4.7 – 551 <LOD – 1.4 0.6 – 141 This study
Marine fish (ww) Nordic countries
<LOD – 0.4 0.9 – 62 <LOD – 5.4 <LOD – 30
This study
The PFAS levels found in the Nordic whale samples (e.g., minke whales, pilot whales) were considerably lower contaminated as previously reported for marine mammals (dolphin liver) from Middle America (Mexico, USA).
57 of 107
5.5. Level comparison with conventional “legacy” persistent organic pollutants Persistent organic pollutants (POP) have been investigated comprehenisively in all environmental contaminants throughout the past decades. Although many of the legacy POPs are banned or restivted in use on a international basis, high levels are still remaining in the environment documenting their high persistence and stability in the environment. Compounds like polychlorinated biphenyls (PCB), hexachlorocyclohexans (HCHs) and hexachlorobenzene (HCB) are still in the focus of environmental concern although banned for decades.
In order to illustrate the environmental significance of our findings in the light of the general contamination of environmental compartments with persistent organic contaminants, a first comparison with levels reported for selected conventional persistent organic pollutants (POP) from the literature is presented (table 20).
5.5.1. Concentration comparison in Seawater
PCBs belong to the most investigated POP compounds worldwide since they were detected for the first time in the 1960s. PFAS levels in the Nordic environmental samples analysed are in general found in comparable concentration ranges as reported for PCB. For seawater the levels found for sum PFAS was considerabley higher than the PCB values reported for the North Sea. Also compared to γ-HCH (a still in-use pesticide) and HCB, the levels for PFAS are considerably higher in North Sea water (table 20).
5.5.2. Concentration comparison in Fresh water
The PFAS levels in fresh water (lake Mjøsa) are also higher than reported for PCBs in Russian rivers. However γ-HCH levels in freshwater are exceeding the Sum PFAS concentration reported for the Nordic countries in average 10 times. The polycyclic musk compounds Galaxolide (HHCB) and Tonalide (AHTN) are found in even higher concentrations in fresh water systems (table 20).
5.5.3. Concentration comparison in rain water
The concentration levels for PFAS found in wet depostion (rain water) is found in the same concentration range than reported for γ-HCH. However, hexachlorobenzene (HCB) was found in considerably higher levels than PFAS. In the newly reported ThemaNord report 503 ”Synthetic musks in the Nordic environment”, rain water samples were analysed for synthetic musks. Synthetic musks were found in the same concentration range as reported for PFAS.
5.5.4. Concentration comparison in sediments
PCBs in contaminated sediments are found in higher concentration ranges than PFAS indicating the importance of considering the different physico-chemical properties of the two compound groups for a proper environmental fate assessment. However, both γ-HCH and HCB are generally reported in concentration ranges comparable to PFAS in Nordic sediments. Synthetic musks (HHCB and AHTN) are reported in concentrations exceeding PFAS levels in several orders of magnitudes (although
58
58
reported on a dry weight basis). PBDE 47 (2,2’,4,4’-TeBDE) is found in considerably lower concentration ranges in Swedish Sediments than found for PFAS in the Nordic samples (table 20).
5.5.5. Concentration comparison in sewage effluent
In landfill effluent, sum PFAS levels from the here presented Nordic study exceed sum PCB concentrations from New Jersey, USA by one order of magnitude (table 20).
5.5.6. Concentration comparison in sewage sludge
Also for conventional POPs, sewage sludge is always a medium where high concentrations can be expected. PCBs levels are exceeding the levels reported for PFAS. Also synthetic musks (HHCB and AHTN) are orders of magnitude higher concentrated in sewage sludge compared with PFAS (Table 21). PBDE levels (PBDE 47) were found to be in a similar concentration range in sewage sludge as reported for PFAS.
5.5.7. Concentration comparison in landfill effluent
In landfill effluent, PCB concentrations reported are found in a similar concentration range (New Jersey, USA) as reported for PFAS in the here presented Nordic study. However, the maximum values for PFAS are about 10 times higher then the PCB concentrations reported in US landfill effluent (waste water).
5.5.8. Concentration comparison in marine mammals
Marine mammals represent a very heterogeneous group of marine animals with regard to trophic levels, food and migration habits as well as physiology and adaptations to their environments. However, in order to provide first indication for compound group specific distribution, this generalisation is thought to be adequate. PFAS levels in marine mammals from Nordic countries (grey-/ harbour seals, pilot whales and minke whales) were found in similar concentration ranges as reported for PCBs, HCB and γ-HCH in marine mammals from Norwegian coastal waters. The concentration reported for PBDE47 is considerably lower than the PFAS levels found (Table 20).
5.5.9. Concentration comparison in marine fish
Concentration ranges reported for conventional POPs in marine fish samples are varying considerably depending on sampling location and species caught. However, in general it was found that PFAS levels in the marine fish species collected during the Nordic screening exercise are lower than reported for PCBs. On the other hand, γ-HCH and PBDE47 are usually reported in lower concentrations than PFAS. HCB is found in a similar concentration range as PFAS.
59 of 107
60
60
5.5.10. Concentration comparison in fresh water fish
PCB and PFAS levels are found in a similar concentration range. However since PCB levels are reported on a lipid weight basis (lw), PCBs in freshwater fish might be even somewhat lower concentrated. Data from Faroe Islands Arctic char (Larsen and Dam 2003) indicate that PCB7 are in the same order of magnitude as reported for PFAS. Synthetic musks (HHCB and AHTN) and brominated flame retardants (PBDE 47) are found in considerably higher levels than reported for PFAS.
5.5.11. Concentration comparison in seabirds
PCB concentrations reported from glaucous gulls (Arctic) are in the same concentration range as found for PFAS in Fulmar eggs from the Faroe Islands (table 21). However PCB7 and HCB were also analysed in the Fulmar eggs used for the PFAS screening study (Mikkelsen et al. 2002). The PCB7 concentrartions exceed the PFAS levels considerably whereas HCB weas int the same concentration range as reported for PFAS (Table 20)
5.6. Toxicity and ecotoxicity Today, very little is known about toxicity and ecotoxicity of PFAS related compounds. Some toxicity data exist for the already identified major environmental PFAS contaminants PFOS and PFOA (Butenhoff et al. 2004). Recently, evidence for neonatal toxicity on Sprague-Dawley rats was reported (Grasty et al. 2003, Thibodeaux et al. 2003). Even invertebrates show toxic effects when exposed to high levels of PFOS (Boudreau et al. 2003). PFOA has been identified as potent peroxisome proliferator in mice (Xie et al. 2003). Genotoxic effects in rates have been postulated for PFOA (Butenhoff et al. 2003). Strong influences on zooplankton populations upon PFOA exposure are reported recently (Sanderson et al. 2003). Hu et al. (2003) found for a comprehensive number of PFAS (incl. PFOS, PFBS, PFHxS) toxic properties leading to alterations in cell membranes. Boudreau et al (2001) reported low LC50 results for PFOA, PFOS and PFDA. There are no scientifically evaluated data reported or published neither on PHpA, PFHpS, PFDS nor on PFNA. A list of reported information on PFOS and PFOA is summarised in table 21.
Compared with the available toxicological and ecotoxicological data, all concentration values are well below the reported threshold levels (Table 20). The reported LC50 and NOEC values are all in the mg-range whereas the PFAS concentrations in the Nordic screening studie are found in the upper ng – lower µg range. However, no information about chronic exposure to low levels is available yet.
More detailed information about tosicology/Ecotoxicology of PFAS related compounds can be found in the references listed in table 21.
61 of 107
62
62
5.7. Sources and environmental implications The here presented Nordic screening exercise on the fate and occurrence of PFAS related residues in the environment confirmed that landfill deposition and sewage treatment are to be considered as important sources for the release of PFAS residues into the aqueous environment (seawater and freshwater). Landfills and waste deposition sites are obviously to be considered as major sources due to the comprehensive application profile of PFAS related products (household related products, textiles, flame retardants etc.). Thus, the Nordic screening on PFAS confirms the first suspicion the waste dumps might play a major role as primary sources for PFAS residues in the environments.
However, measurable amounts of PFAS were found also in precipitation samples from Sweden and Finland demonstrating, that selected PFAS chemicals are prone to atmospheric transport and sub-sequent wet deposition via rain. Thus, selected PFAS related contaminants should be evaluated with regard to long-range transport potential.
PFOS and PFOSA dominated the PFAS burden in top predating organisms and thus indicate strong bioaccumulation tendencies. These findings support earlier reports where PFOS and PFOSA were detected in high concentrations in top predating species even from Arctic regions. PFAS related compounds like PFOS and PFOSA must therefore be evaluated as persistent and bio-accumulative residues.
These basic properties characterising many PFAS related chemicals as persistent and bioaccumulative in conjunction with recent ecotoxicological studies supports the need to add PFAS into the UNEP category of PBT compounds: “Persistent, bioaccumulative and toxic”.
5.8. Consequences and recommended actions The here presented report contributes to the international scientific efforts to elucidate the hazarous potential of PFAS-related contaminants in the environment. The results presented illustrate the heterogeneity of this group of contaminants with regard to physico-chemical properties, accumulation potential and ecotoxicological potential.
PFAS related contamination was detected in all samples from all participating Nordic countries. These contaminants are present in all countries and, thus, an ubiquitous distribution must be assumed not only across the Nordic countries but globally.
Compounds with considerable bioaccumulation potential like PFOS and PFOSA are found in high concentrations in top predating marine mammals like harbour and grey seals as well as in pilot whales.
As earlier described, PFAS residues are practically not biodegradable and are, thus, expected to be present in the environment for many decades to come. It is therefore recommended to include these PFAS related chemicals major environmental hazards in the national and international monitoring programs for environmental contaminants.
63 of 107
6. Conclusions and suggestions
The here presented results from a first screening on perfluorinated alkylated substances (PFAS) is contributing to the international scientific efforts to evaluate and characterise the environmental behaviour of this very special group of environmental contaminants. A core group of 6 PFAS compounds PFOS, PFOA, PFOSA, PFHxA, PFNA, PFHxS) was analysed in all 119 samples covering seawater, rain water, fresh water, sewage sludge, sewage effluent, landfill effluent, seabird eggs (Fulmar) and liver samples from 16 species of marine and freshwater fish as well as marine mammals. As a voluntary addition to the contracted work, PFHpA (perfluoroheptanoic acid), PFBS (perfluorobutane sulfonate) and PFDS (perfluorodecane sulfonate) were analysed.
PFAS residues were found in all environmental matrices analysed. Specific matrix- dependent distribution patterns were found. PFOS and PFOA were usually dominating in abiotic samples whereas PFOS and PFOSA were the predominant PFAS residues in biota. For abiotic samples, the highest PFAS levels were found in landfill effluent and sewage sludge whereas the top-predating grey seals were highest contaminated in the biota sample set.
The PFAS levels found were usually in the same order of magnitude or even higher than reported for conventional “legacy” persistent organic pollutants. In sewage samples, however, PCBs and synthetic musks were higher concentrated.
Until today, very little information is available on the ecotoxicological behaviour and effects of PFAS in biological systems. However, the concentrations found in landfill effluents and sewage are detected in concentration ranges where chronic exposure might cause adverse effects in ambient biosystems.
Based on the here presented results, PFAS compounds are identified as chemicals with considerable environmental distribution. Thus, the major contributors of this contaminant group should be implemented in on-going monitoring activities after analytical methods are adapted to this purpose. Due to their physico-chemical properties, most of the PFAS compounds are not accumulating in lipid-rich tissues. Thus, matrices bird eggs, like liver, bile and blood should be preferred for monitoring of PFAS. It is recommended to include freshwater fish located close to possible waste dumps since such locations may represent major sources for environmental distribution of PFAS. Long-range transport as a possible source should be included in futiure environmental monitoring planning. Thus, more in-depth sampling in coastal waters and remote atmosphere should give information on the long-range transport potential.
PFAS residues are virtually indestructible and, thus, will eventually reach concentrations in biota with effects on physiology and environmental health.
64
64
7. Acknowledgements
The support and help of many co-workers is greatly acknowledged. Without the help of numerous colleagues from the respective Nordic institutions responsible for sampling and sample handling, the project would not have been possible to realise.
Sample preparation, analysis and quantification: Arve Bjerke (NILU) was responsible for preparation of the abiotic samples. We appreciate the help of Christian Dye (NILU) during analysis and quantification of the abiotic samples. Katrin Holmström (ITM) and Carina Johansson (ITM) assisted Ulf Järnberg (ITM) in the collecting and preparation of the biological samples.
Sample collection: The following colleagues have been responsible for/ supported the study for the national sampling for the Nordic screening for PFAS related compounds in the environment. In addition numerous institutions/ organisation contributed with help, advise and personnel to the success of the here presented study:
Norway: Eirik Fjeld, Norwegian Institute of Water Research (NIVA) provided water and biota samples, Trine Eggen, Norwegian Centre for Soil and Environmental Research (Jordforsk) provided landfill and sludge samples.
Denmark: The counties of Copenhagen, Roskilde, Fyn and Ringkøbing (water, sludge mussels and fish) contributed to the sample collection. Martin Larsen, National environmental Research Institute NERI) maintained the contact to the county officials. Jonas Theilman, (NERI) is acknowledged for providing the harbour seal liver samples. Finland: Sirpa Paattakainen (rainwater sampling) and Tarja Bertula from SYKE's Laboratory performed the pike sample dissection. The Uusimaa Regional Environment Centre was responsible for seawater and sediment sampling, Helsinki City provided sludge as well effluent samples and performed pike fishing. Espoo City and Porvoo City provided sludge and effluent samples and Helsinki Metropolitan Area Council was contributing with landfill effluent sampling.
Iceland: In Iceland the following specialists helped to organise sampling of specimens and materials for the PFOS analysis: Gudjun Atli Audunsson, The Icelandic Fisheries Laboratories and Jörundur Svavarsson, Institute of Biology, University of Iceland. Faroe Islands: In the Faroe islands the following specialists and associations have contributed during sampling and sample handling: Katrin Hoydal and Jóhannis Danielsen (marine fish
65 of 107
sampling), the Faroe Islands Angelers Association, as well as Jens-Kjeld Jensen, Poul Jóhannis Simonsen og Janus Hansen. Sweden: The follwoing experts have contributing to sample collection and handling: Mikael Svensson, MS Naturfakta, Osby: Abborre Helgeån Kristianstad Olle Viberg, Simrishamn: Atlantic cod (Hanöbukten), Tjelvar Odsjö and coworkers, Gruppen för miljögiftsforskning, Naturhistoriska Riksmuseet: Atlantic cod from Hoburgen (SE off Gotland): Charlotta Persson and Kristianstads municipality (STP slam).
66
66
8. Literature
AMAP (1998) AMAP Assessment Report: Arctic pollution issues. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway. XII + pp.859.
AMAP (2004) AMAP assessment 2002: Persistent Organic Pollutants in the Arctic. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway. XVI + pp. 310.
Anonymous (1999) The science of organic fluorochemistry. 3M Report OPPT-2002-0043-0006: pp. 12.
Anonymous (2000) Naturmiljön i siffror 2000. Swedish Environmental Protection agency, pp. 296.
Anonymous (2002) Environmental directorate, Joint meeting of the chemicals committee and the working party on chemicals, pesticides and biotechnology. Co-operation on existing chemicals. Hazard assessment of perfluorooctane sulfonate (PFOS) and its salts. Report JT00135607; ENV/JM/RD(2002)17/FINAL: pp. 362.
Berg T., Kallenborn R., Manø S. & Uggerud H. Th. (2003) Tidstrender i atmosfærekonsentrasjoner av tungmetaller og persistente organiske miljøgifter. Statlig Program for Forurensningsovervåking (SFT), report 883/03, pp. 24
Bernthoft A. & Skaare J.U. (1994) Levels of selected individual polychlorinated biphenyls in different tissues of harbour seals (Phoca vitulina) from the southern coast of Norway. Environ. Poll. 86:99-107
Boudreau T.M. Janutka R., Solomon K.R., Muir D.C.G. & Mabury S.A. (2001) Perfluorinated surfactants; Aquatic Toxicity studies on freshwater primary and secondary trophic levels. SETAC Europe conference, Madrid (Spain).
Boudreau T.M., Wilson C.J., Cheong W.J., Sibley P.K., Mabury S.A., Muir D.C.G & Solomon K.R. (2003) Response of the zooplankton community and environmental fate of perfluorooctane sulfonic acid in aquatic microcosms. Environ. Toxicol. Chem. 22/11: 2739-2745.
Brook. D., Footitt A. & Nwaogu T.A. (2004) Environmental risk evaluation report: Perfluorooctanesulfonate (PFOS). UK Environmental Agency, Isis House, Howbery Park, Wallingford OX10 8BD, in preparation.
Buegel-Mogensen B., Pritzl G., Rastogi S., Glesne O., Hedlund B., Hirvi J.-P., lundgren A. & Sigurdsson A. (2004) Synthetic musks in the Nordic environment. ThemaNord 2003-503, Nordic Council of Ministers, Copenhagen 2004, ISBN 92-893-0981-4: pp.69.
Butenhoff J.L., Kennedy G.L., O'Connor J.C. & York R.G. (2003) Multi-generation reproduction study of ammonium perfluorooctanoate in rats. Toxicologist 72/S-1: 76.
Butenhoff J.L., Gaylor D.W., Moore J.A., Olsen G.W., Rodricks J. Mandel J.H. & Zobel L.R. (2004) Characterization of risk for general population exposure to perfluorooctanoate. Regul. Toxicol. Pharmacol. 39: 363-380.
67 of 107
Cederberg I., Edell Å. & Lindqvist I. (2004) PFOS; Rapport från en förstudie. Report, kemikalieinspektionen, pp.21.
Chow S.C. (2003) Encyclopedia of Biopharmaceutical Statistics, 2nd Edition. Dekker, pp. 1000 , ISBN: 0-8247-4261-3.
Falandysz J., Kannan K. Tanabe S. & Tatsukava R (1994) Organochlorine pesticides and polychlorinated biphenyls in cod-liver oils: North Atlantic, Norwegian Sea, North Sea and Baltic Sea. Ambio 23/4-5: 288-293.
Giesy J. P. & Kannan, K. (2001) Global distribution of perfluorooctane sulfonate in wildlife. Environ. Sci. Technol. 35 : 1339-1342.
Grasty R.C., Grey B.E., Lau C.S. & Rogers J.M. (2003) Prenatal window of susceptibility to perfluorooctane sulfonate-induced neonatal mortality in the Sprague-Dawley rat. Birth Defects Res. Part B Dev. Reprod. Toxicol. 68/6: 465-471.
Green N., Helland A, Hylland K., Knutzen J & Walday M. (2001) Joint Assessment and Monitoring Programme (JAMP), Overvåking av miljøgifter i marine sedimenter og organismer 1981 – 1999. Statlig Program for forurensingsovervåking. Overvåkingsrapport nr. 819/01 TA-nr. 1797/2001: pp. 191.
Knutzen J & Green N. (2001) Joint Assessment and Monitoring Programme (JAMP), ”Bakgrunnsnivåer” av miljøgifter i fisk og blåskjellbasert på datamaterialet fra 1990 – 1998. Statlig Program for forurensingsovervåking. Overvåkingsrapport nr. 820/01 TA-nr. 1798/2001: pp. 145.
Haglund P., Zoog D.R., Buser H.-R., & Hu J. (1997) Identification and quantification of polybrominated diphenyl ethers and methoxy-polybrominated diphenyl ethers in Baltic biota. Env. Sci. Technol. 31 : 3281-3287.
Hansen, K. J., Clemen, L. A., Ellefson, M. E. & Johnson, H. O. (2001) Compound-specific, quantitative characterization of organic fluorchemicals in biological matrices. Environ. Sci. Technol. 2001, 35, 766-770.
Hansen K.J ., Johnson H.O., Eldridge J.S., Butenhoff J.L. & Dick L.A. (2002) Quantitative characterization of trace Levels of PFOS and PFOA in the Tennessee River. Environ. Sci. Technol.36, 1681-1685
Hekster F.M., de Voogt P., Pijnenburg A.M.C.M.& Laane R.W.P.M. (2002) Perfluoroalkylated substances ; Aquatic environmental assessment. Rijksinstituut voor Kust en Zee, Report RIKZ/2002.043:
Henriksen E.O., Gabrielsen G.W. & and Skaare J.U. (1996) Levels and congener pattern of polychlorinated biphenyls in kittiwakes (Rissa tridactyla), in relation to mobilization of body lipids associated with reproduction. Environ. Poll. 92/1: 27-37.
Herbert G.N., Odom M.A., Craig P.S., Dick D.L. & Strauss S.H. (2002) Method for the determination of sub-ppm concentrations of perfluoroalkylsulfonate anions in water. J. Environ. Monit. 2002, 4, 90-95.
Hoff P.T., Van de Vijer K., Van Dongen W., Esmans E.L:, Blust R. & DeCoen W. (2003) Perfluorocctane sulfonic acid in bib (Tripterus luscus) and plaice (Pleuronectes platessa) from the Western Scheldt and the Belgian North Sea: distribution and biochemical effects. Environ. Toxicol. Chem. (2003) 22, 608-614.
68
68
Holte B., Næs K., Dahle S. & Gulliksen B. (1994) Marine resipientundersøkelser ved Longyearbyen og Barentsburg, Svalbard. Bunndyrssamfunn og miljøgifter i bunnsedimenter. Akvaplan-niva rapport 412.94.402. pp. 47.
Hu W., Jones P.D., DeCoen W., King L., Fraker P., Newsted J. & Giesy J.P. (2003) Alterations in cell membrane properties caused by perfluorinated compounds. Comp Biochem Physiol C Toxicol Pharmacol. 135/1:77-88
Kallenborn R., Gatermann, G., Nygard T., Knutzen J. & Schlabach M. (2001) Synthetic musks in Norwegian marine fish samples collected in the vicinity of densely populated areas. Fres. Environ. Bull. 10/11: 832-842
Kannan, K., Koistinen, J., Beckmen, K., Evans, T., Garzelany, J., Hansen, K. J., Jones, P. D. & Giesy, J. P. (2001a) Accumulation of Perfluorooctane Sulfonate in Marine Mammals. Environ. Sci. Technol. 35: 1593-1598.
Kannan, K., Hansen, S. P., Franson, C. J., Bowerman, W. W., Hansen, K. J., Jones, P. D. & Giesy, J. P. (2001b) Perfluorooctane sulfonate in fish-eating water birds including bald eagles and albatrosses. Environ. Sci. Technol. 35, 3065-3070.
Kannan, K., Newsted, J. L., Halbrook, R. S. & Giesy, J. P. (2002a) Perfluorooctanesulfonate and related fluorinated hydrocarbons in mink and river otters from the United States. Environ. Sci. Technol., 36: 2566-2571.
Kannan, K., Corsolini, S., Falandysz, J., Oehme, G., Focardi, S. & Giesy, J. P. (2002b) Perfluorooctanesulfonate and Related Fluorinated Hydrocarbons in Marine Mammals, Fishes, and Birds from Coasts of the Baltic and the Mediterranean Seas. Environ. Sci. Technol. 2002, 36, 3210-3216.
Kannan K., Choi J.W., Iseki N., Senthilkumar K., Kim D.H., Masunaga S. &bGiesy J.P. Concentrations of perfluorinated acids in livers of birds from Japan and Korea (2002c) Chemosphere 49/3: 225-231
Kannan K., Hansen K.J., Wade T.L.& Giesy J.P. (2002d) Perfluorooctane sulfonate in oysters, Crassostrea virginica, from the Gulf of Mexico and the Chesapeake Bay, USA. Arch. Environ. Contam. Toxicol. (2002d) 42, 313-318.
Kjemikalieinspektionen (2004) PFOS-relaterade ämnen. Strategi för utfasning. Inkludert Bilag 3: riskbedömning av PFOS. Kemikalieinspeskjonen (Sweden) pp. 85.
Kleivane L., Skaare J.U., Bjørge A., de Ruiter E. & Reijnders P.J.H. (1994) Organochlorine pesticide residue and PCBs in harbour porpoise (Phocoena phocoena) incidentally caught in Scandinavian water. Environ. Poll. 89/2: 137-146.
Lange F .T ., Schmidt C.K., Metzinger M., Wenz M. & Brauch H.-J. (2004) Determination of perfluorinated carboxylates and sulfonates from aqueous samples by HPLC-ESI-MS-MS and their occurrence in surface waters in Germany. poster presentation MOPO15/006, SETAC Europe conference, Prague
Larsen R.B. & Dam M. (2003) In: Hoydal and Dam (Eds.) AMAP Greenlanbd and the Faroe Islands 1997 – 2001. DANCEA, Ministry of Environment. Pp. 265.
Law R.J., Stringer R.L., Allchin C.R. & Jones B.R. (1996) Metals and organochlorines in sperm whale (Physeter macrocephalus) stranded around the North Sea during 1994/1995 winter. Marine Poll. Bull. 32/1: 72-77
69 of 107
Levitt D. & Liss A. (1986) Toxicity of perfluorinated fatty acids for human and murine B cell lines. Toxicol Appl Pharmacol. 86/1: 1-11.
Martin J.W., Smithwick M.M., Braune B.M., Hoekstra P.F., Muir D.C.G. & Marbury S.A. (2004) Identification of long-chain perfluorinated acids in biota from the Canadian Arctic. Environ. Sci. Technol 38: 373-380.
McNaught A.D. & Wilkinson A. (1997), IUPAC Compendium of Chemical Terminology, 2nd Edition, Blackwell Science, 1997 [ISBN 0-86542-6848].
Moody C.A. & Field J. (2000) Perfluorinated surfactants and the environmental implications of their use in fire-fighting foams. Environ. Sci. Technol. 34, 3864-3870.
Mikkelsen B., Hoydal K., Dam M., & Danilesen J. (2002) Føroya Umhvørvi í Tølum, 2001. Heilsufrøδiliga Starvstovan, 2001: pp. 130.
Olson C.T. & Andersen M. E. (1983) The acute toxicity of perfluorooctanoic and perfluorodecanoic acids in male rats and effects on tissue fatty acids. Toxicol Appl Pharmacol 70(3):362-372.
Olsen G.W., Hansen K.J., Stevenson L.A., Burris J.M. & Mandel J.H. (2003). Human donor Liver and Serum concentrations of perfluorooctanesulfate and other perfluorochemicals. Environ. Sci. Technol. 37: 888-891.
Renner R., (2001) Growing Concern over Perfluorinated Chemicals. Environ. Sci. Technol.35/7 : 154A.
Rimkus G.G. volume Ed. (2004) Synthetic musks in the Environment. In: Hutzinger O. (Editor in Chief) The Handbook of Environmental Chemistry. Springer Vlg. ISBN 3-540-043706-1. pp.338.
Samsøe-Petersen L. (2003) Organic contaminants in sewage sludge. Swedish Environmental Protection agency, Report 5217. pp. 71.
Sanderson H., Boudreau T.M., Mabury S.A. & Solomon K.R. (2003) Impact of perfluorooctanoic acid on the structure of the zooplankton community in indoor microcosms. Aquat Toxicol. 62(3):227-34.
Sapota G. (2004) Polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs) in seawater of the southern Baltic Sea. Desalination 162: 153–157. Schrap S.M., Voogt, de P., van Leeuwen S.P., Pijnenburg A.M.C.M..(2004) Perfluorated compounds in the Dutch aquatic environment. poster presentation MOPO15/008, SETAC Europe conference, Prague Schrock M., Tracy K., Misita M., Tabor J., Durell G., Dahlen D., Liu B., DeGraeve M.,& McCauley D. (2003) Wastewater contaminants measured to support the New Jersey Toxics reduction program – Part 2: dioxin, PCB and organochlorine pesticide profiles. Organohalogen Comp. 62:
Schröder H. F. (2003) Determination of fluorinated surfactants and their metabolites in sewage sludge samples by liquid chromatography with mass spectrometry and tandem mass spectrometry after pressurised liquid extraction and separation on fluorine-modified reversed-phase sorbents. J. Chrom. A (2003), 1020: 131-151
Schulz-Bull D., Petrick G & Duinker J.C. (1991) Polychlorinated biphenyls in North Sea water. Marine Chemistry (1991) 365-384.
70
70
Sellström U., Kierkegard A., deWit C. & Jansson B. (1998) Polybrominated diphenyl ethers and hexabromocyclododecane in sediment and fish from a Swedish river. Environ. Toxicol. And. Chem. 17/6: 1065-1072.
Taniyasu S., Kannan K., Horii Y., Hanari N. & Yamashita N. (2003). A survey of perfluorooctane sulfonate and related perfluorinated organic compounds in water fish, birds and humans from Japan. Environ. Sci. Technol. 37: 2634-2639.
Theobald N, Gaul H. & Zierbarth U (1996) Verteilung von organischen Schadstoffen in der Nordsee und angrensenden Seegebieten. Deutsche Hydrographische Zeitschrift, Supplement 6.: 81-93.
Thibodeaux J., Hanson R.G., Grey B.E., Barbee B.D., Richards J.H., Butenhoff J.L., Rogers J.M. & Lau C. (2003) Maternal and developmental toxicity of perfluorooctane sulfonate (PFOS) in the mouse. Toxicologist 2(S-1): 342
Thomsen V., Schatzlein D & Mercuro D. (2003) Limits of detection in spectroscopy. Spectroscopy 18/12: 112-114.
U.S. Environmental Protection Agency (2000) Perfluorooctyl Sulfonates: Proposed Significant New Use Rule. Fed. Regist. 65: 62319-62333.
U.S. Environmental Protection Agency (2003) Preliminary risk assessment of the development toxicity associated with exposure to perfluorinated octanoic acid. Rreport. pp. 63.
Van de Vijer K. I., Hoff P.T., Das K., van Dongen W., Esmans E.L., Jauniaux T., Buoquegneaux J.-M., Blust R.& deCoen W. (2003) Perfluorinated chemicals infiltrate Ocean Waters: Link between Exposure Levels and Stable Isotope Ratios in Marine Mammals. Environ. Sci. Technol. 37: 5545-5550.
Xie Y., Yang Q., Nelson B.D. & DePierre J.W. (2003) The relationship between liver peroxisome proliferation and adipose tissue atrophy induced by peroxisome proliferator exposure and withdrawal in mice. Biochem Pharmacol. 66/5:749-756.
Ylinen,M., Hanhijarvi, H., Peura,P., & Ramo O. (1985) Quantitative gas chromatographic determination of perfluorooctanoic acid as the benzyl ester in plasma and urine. Arch Environ Contam Toxicol, 14: 713-717
3M (2000) Sulfonated Perfluorochemicals in the Environment: Sources, Dispersion, Fate and Effects. Report pp. 51
71 of 107
Appendix 1: Sample characteristics
Biota Sample ID Country/ Location Species Sampling position Sampling date Sample type Comments
Denmark
FLDAN01 Köbenhavn Amt Flounder 55.44 º N – 12.36 º E Liver
FLDAN02 Köbenhavn Amt repl. Flounder 55.44 º N – 12.36 º E Liver
FLDAN03 Köbenhavn Amt repl. Flounder 55.44 º N – 12.36 º E Liver
FLDAN04 Köbenhavn Amt repl. Flounder 55.44 º N – 12.36 º E Liver
EPDAN05 Roskilde fjord Eelpout 55.44 º N – 12.01º E Liver
HEDAN06 Roskilde fjord Herring 55.44 Liver
HSDAN07 Limfjord1 Harbour seal 56.40º N - 08.20º E Liver
HSDAN08 Limfjord2 Harbour seal 56.40º N - 08.20º E Liver
HSDAN09 Hesselö Harbour seal 56.10º N - 11.40º E Liver
HSDAN10 Samsö Harbour seal 55.50º N - 10.30º E Liver
HSDAN11 Öresund Harbour seal 55.35º N - 12.50º E Liver
Finland
PIFIN01 Helsinki City coast Pike 60.11° N – 24.59° E Liver
72
72
Sample ID Country/ Location Species Sampling position Sampling date Sample type Comments
PIFIN02 Helsinki City coast Pike 60.11° N – 24.59° E 2003 Liver Vanhankaupungin lahti; Gammelstadsfjärden
PIFIN03 Helsinki City coast Pike 60.11° N – 24.59° E 2003 Liver Vanhankaupungin lahti; Gammelstadsfjärden
PIFIN04 Helsinki City coast Pike 60.11° N – 24.59° E 2003 Liver Vanhankaupungin lahti; Gammelstadsfjärden
PIFIN05 Helsinki City coast Pike 60°11’N - 24°59’ E 2003 Liver Vanhankaupungin lahti; Gammelstadsfjärden
PIFIN06 Helsinki City coast Pike 60°11’N - 24°59’ E 2003 Liver Vanhankaupungin lahti; Gammelstadsfjärden
PIFIN07 Helsinki City coast Pike 60°11’N - 24°59’ E 2003 Liver Vanhankaupungin lahti; Gammelstadsfjärden
PIFIN08 Helsinki City coast Pike 60°11’N - 24°59’ E 2003 Liver Vanhankaupungin lahti; Gammelstadsfjärden
Faroe Islands
SCFAR01 Kaldbak Sculpin 62.03 N – 06.49° W May 2002 Liver
SCFAR02 Kaldbak Sculpin 62.03 N – 06.49° W May 2002 Liver
DAFAR03 Kaldbak Dab 62.03 N – 06.49° W July 2002 Liver
DAFAR04 Kirkjubö Dab 61.57° N – 06.47° W July 2002 Liver
COFAR05 Torshavn Cod 62.00° N – 06.46° W June 2002 Liver
ACFAR06a á Mýranar, Bleiki I Arctic char 62.10° N – 07.05° W July 2002 Liver Landlocked
ACFAR06b á Mýranar , Bleiki I Arctic char 62.10° N – 07.05° W July 2002 Liver Landlocked
73 of 107
ACFAR07 á Mýranar, Bleiki II Arctic char 62.10° N – 07.05° W July 2002 Liver Landlocked
PWFAR09 Sandagerdi Pilot whale 62.00° N – 06.46° W September 2002 Liver
Sample ID Country/ Location Species Sampling position Sampling date Sample type Comments
PWFAR10 Sandagerdi Pilot whale 62.00° N – 06.46° W September 2002 Liver
PWFAR12 Sandagerdi Pilot whale 62.00° N – 06.46° W September 2002 Liver
PWFAR13 Sandagerdi Pilot whale 62.00° N – 06.46° W September 2002 Liver
FUFAR14 Vidareidi Fulmar pool1 62.21° N – 06.30° W 21.05.2002 egg Pooled samples
FUFAR15 Vidareidi Fulmar pool2 62.21° N – 06.30° W 21.05.2002 egg Pooled samples
Iceland
LDICE01a Gufunes bay Long rough dab
64.09º N, 21.48º W 17.10.2003 Liver
LDICE01b Gufunes bay Long rough dab
64.09º N, 21.48º W 17.10.2003 Liver
LDICE01c Gufunes bay Long rough dab
64.09º N, 21.48º W 17.10.2003 Liver
LDICE02 Gufunes bay Long rough dab
64.09º N, 21.48º W 18.10.2003 Liver
SCICE03 Videyarbryggja Sculpin 64.09º N, 21.51º W 09.09.2003 Liver
DAICE04 Gufunes bay Dab 64.09º N, 21.51º W 17.10.2003 Liver
MWICE05 Atlantic ocean Minke whale 66.10º N, 19.16º’ W 26.08.2003 Liver
MWICE06 Atlantic ocean Minke whale 66.12º N, 14.39º’ W 30.08.2003 Liver
MWICE07 Atlantic ocean Minke whale 63.50º N, 22º 49’ W 27.08.2003 Liver
MWICE08 Atlantic ocean Minke whale 64.36º N, 22.51º W 30.08.2003 Liver
74
74
MWICE09 Atlantic ocean Minke whale 64.13º N, 22.39º W 15.09.2003 Liver
Sample ID Country/ Location Species Sampling position Sampling date Sample type Comments
Norway
BUNOR01 Mjösa: North of Gjøvik Burbot 60.82º N – 10.67º E Juni 2003 Liver
TRNOR02 Mjösa, central basin Trout 60.72º N – 10.96º E Juni-August 2003 Liver Pooled samples
PINOR03 Mjösa, North of Gjøvik Pike 60.82º N – 10.67º E Mai-June 2003 Liver
Pooled samples
PENOR04 Mjøsa, North of Gjøvik Perch 60.82º N – 10.67º E 15.06.2003 Liver
Sweden
PESWE01 Helgeån Gummastorpasjäön Perch 56.00º N - 12.10º E Liver
PESWE02 Helgeån Araslövssjön Perch 62.15º N – 13.95º E Liver
PESWE03a Helgeån Hammarsjön Ö repl. Perch 62.10º N – 13.99º E Liver
PESWE03b Helgeån Hammarsjön Ö repl. Perch 62.10º N – 13.99º E Liver
PESWE04 Helgeån Hammarsjön N Perch 62.08º N – 14.05º E Liver
COSWE05 Hoburgen Cod 56.54º N - 18.10º E Liver
COSWE06 Hoburgen Cod 56.54º N - 18.10º E Liver
75 of 107
COSWE07 Hoburgen Cod 56.54º N - 18.10º E Liver
COSWE08 Hoburgen Cod 56.54º N - 18.10º E Liver
COSWE09 Hoburgen Cod 56.54º N - 18.10º E Liver
COSWE10 Hoburgen Cod 56.54º N - 18.10º E Liver
Sample ID Country/ Location Species Sampling position Sampling date Sample type Comments
COSWE11 Hanöbukten Cod 55.41º N -14.21º E Liver
COSWE12 Hanöbukten Cod 55.41º N -14.21º E Liver
GSSWE13 Sörmland Grey seal 58.50º N - 18.00º E Liver
GSSWE14 Uppland Grey seal 60.30º N - 18.20º E Liver
GSSWE15 Gästrilkand Grey seal 61.00º N - 17.50º E Liver
76
76
Seawater, freshwater, Sediment and sludge samples
Sample type Sample ID Sampling locationSample amount analysed Sampling date/ time Additional information Position: Lat. –Long.
Sea water SWFIN01 Helsinki 497g 21.08.2003, 10.00 h Brackish sea water, main city area
60.11º N’ - 24.59º E
Sea water SWFIN02 Helsinki 517g 21.08.2003, 13.00 h Outside Helsinki 60.06º N - 24.45º E
Sea water SWFIN03 Porvoo 490g 19.08.2003, 11.00 h Brackish sea water 60.44º N - 25.33º E
Sea water SWFIN04 Porvoo 503g 19.08.2003, 13.00 h Brackish sea water 60.15º N - 25.32º E
Rain water RWFIN05 Helsinki 492g 23.-27.08.2003 Urban + industrial influences 60.27º N - 24.87º E
Rain water RWFIN06 Helsinki 490g 21.-22.08.2003 Urban + industrial influences 60.27º N - 24.87º E
Landfill effluent water (sub-samples)
LFFIN07a Espoo 492g
02.09.2003 untreated effluent water 60.12º N - 24.32 º E
Landfill effluent water (sub-samples)
LFFIN07b Espoo 496g
02.09.2003 untreated effluent water 60.12º N - 24.32 º E
Landfill effluent water (sub-samples)
LFFIN07c Espoo 497g
02.09.2003 untreated effluent water 60.12º N - 24.32 º E
Sewage effluent water SEFIN08 Helsinki 504g 29.08.2003 Urban + industrial influences 60.27º N - 24.87º E
Sewage effluent water (sub-samples)
SEFIN09a Porvoo 513g
29.08.2002 Urban + industrial influences 60.37º N - 25.61º E’
Sewage effluent water (sub-samples)
SEFIN09b Porvoo 510g
29.08.2003 Urban + industrial influences 60.37º N - 25.61º E’
77 of 107
Sediment (sub-samples) SDFIN10a Helsinki 4.8g 21.08.2003 Site 1 60.06º N- 24.45º E
Sample type Sample ID Sampling locationSample amount analysed Sampling date/ time Additional information Position: Lat. –Long.
Sediment (sub-samples) SDFIN10b Helsinki 5.0g 21.08.2003 Site 1 60º 06’ N- 24º 45’ E
Sediment (sub-samples) SDFIN10c Helsinki 5.0g 21.08.2003 Site 1 60º 06’ N- 24º 45’ E
Sediment SDFIN11 Porvoo 1.6g 19.08.2003, 11.00 h Site 2 60.14º N - 25.33º E
Sediment SDFIN12 Porvoo 2.1g 19.08.2003, 13.00 h Site 1 60.15º N - 25.32º E
Sewage sludge SSFIN13 Espoo 5.2g 01.09.2003 dewatered 60.11º N - 24.74º E
Sewage sludge (sub-samples)
SSFIN14a Helsinki 5.3g
29.08.2003 dewatered 60.27º N - 24.87º E
Sewage sludge (sub-samples)
SSFIN14b Helsinki 5.0g
29.08.2003 dewatered 60.27º N - 24.87º E
Sewage sludge (sub-samples)
SSFIN14c Helsinki 5.0g
29.08.2003 dewatered 60.27º N - 24.87º E
Sewage sludge SSFIN15 Porvoo 4.8g 28.08.2003 dewatered 60.37º ’N - 25.61º E
Sewage sludge SSFIN16 Porvoo 4.3g 19.08.2003 dewatered 60.37º ’N - 25.61º E
Rainwater (sub-samples) RWSWE01a Råö 496g
02.10.2003-11.11.2003
57.39º N - 11.91º E
Rainwater (sub-samples) RWSWE01b Råö 511g
02.10.2003-11.11.2003
57.39º N - 11.91º E
Rainwater RWSWE02 Råö 519g
11.11.2003-03.12.2003
57.39º N - 11.91º E
Rainwater RWSWE03 Råö 515g
03.12.2003-15.12.2003
57.39º N - 11.91º E
Sediment SDSWE04 Kristianstad, 5.2g 23.09.2003 Accumulation site 62.15º N – 13.95º E
78
78
Arnslövssjön
Sample type Sample ID Sampling locationSample amount analysed Sampling date/ time Additional information Position: Lat. –Long.
Sediment SDSWE05 Kristianstad, Hammarsjön II 4.0g
23.09.2003 Accumulation site 56.00º N – 14.15º E
Sediment SDSWE06 Kristianstad, Hammarsjön I 4.5g
23.09.2003 Accumulation site 56.00º N – 14.15º E
Sewage sludge SSSWE07 Kristianstad, Centrala ARV 4.5g
week 39, 2003 collected over the entire week
56.00º N – 14.15º E
Sewage sludge SSSWE08 Köpinge ARV 3.0g
week 39, 2003 collected over the entire week
55.55º N – 14.15º E
Sewage sludge SSSWE09 Tollarp ARV 3.7g
week 39, 2003 collected over the entire week
55.55º N – 14.00º E
Lake water LWNOR01 Lillehammer 504g 01.07.2003 Lake Mjøsa 61.09º N – 10.45º E
Lake water LWNOR02 Gjøvik 501g 01.07.2003 Lake Mjøsa 61.78º N – 10.71º E
Lake water LWNOR03 Furnesfjorden 501g 01.07.2003 Lake Mjøsa 61.87º N – 10.93º E
Lake water LWNOR04a Hamar 515g 01.07.2003 Lake Mjøsa 61.78º N – 11.06º E
Lake water LWNOR04b Hamar 526g 01.07.2003 Lake Mjøsa 61.78º N – 11.06º E
Landfill effluent LFNOR05a Spillhaug 481g 29.09.2003, 06.45 h 59.83º N – 10.85º E
Landfill effluent LFNOR05b Spillhaug 479g 29.09.2003, 06.45 h 59.83º N – 10.85º E
Landfill effluent LFNOR06 Støleheia 483g
26.09.2003 Effluent from komposting treatment 58.25º N - 07.90º E
Landfill effluent LFNOR07 Gålås 528g 02/07/2003 59.92º N – 11.60º E
79 of 107
Landfill effluent LFNOR08 Røyken 499g 27/08/2003 Coarse waste 59.70º N – 10.47º E
Landfill effluent LFNOR09 Spillhaug 497g 18/09/2003 effluent from cleaning plant 59.83º N – 10.85º E
Sewage effluent SENOR10 Lillehammer 521g 01.07.2003 at lake Mjøsa 61.09º’N – 10.45º E
Sample type Sample ID Sampling locationSample amount analysed Sampling date/ time Additional information Position: Lat. –Long.
Sewage effluent SENOR11a Gjøvik 482g 01.07.2003 purified outlet 61.78º N – 10.71º E
Sewage effluent SENOR11b Gjøvik 503g 01.07.2003 purified outlet 61.78º N – 10.71º E
Sediment SDNOR12 Furnes 5.5g 02.09.2003 Lake Mjøsa 60.86º N – 10.92º E
Sediment SDNOR13 Hamar 5.5g 01.09.2003 Lake Mjøsa 60.78º N – 11.06º E
Sediment SDNOR14 Gålås 3.5g 02.07.2003 Lake Mjøsa 60.78º N – 10.71º E
Sewage sludge SSNOR15 Gjøvik 4.9g
week 36, 2003 collected over the entire week
60.78º N – 10.71º E
Sewage sludge SSNOR16 Lillehammer 5.1g
week 36, 2003 collected over the entire week
61.09º N – 10.45º E
Sea water SWDAN01 Limfjorden (Ringkøpingamt) 512g 12.11.2003
56.34º N – 08.29º E
Sea water (sub-samples) SWDAN02a Odensefjord 503g 24.10.2003 S1: Seden strand 55.26º’N - 10.25º’ E
Sea water (sub-samples) SWDAN02b Odensefjord 497g 24.10.2003
55.26º’N - 10.25º’ E 55º 26’N - 10º 25’
Sea water SWDAN03 Roskilde 494g 30.10.2003 Station 60 55.26º N - 12.04º E
in cooperation with Ulf Järnberg, Institute of Applied Environmental Research (ITM)
Stockholm University, SE-106 91 Stockholm , Sweden
83 of 107
1. Introduction and objectives of the study These guidelines concern the sampling, sample handling and shipping of water, sediment, sludge, fish liver and seal liver for trace analysis of organic contaminants. They are suitable for perfluoroalkylated substances (PFAS) such as perfluoroalkylated acids, perfluoroalkylated sulfonates and their derivatives. The institutes performing the chemical trace analysis do not take any responsibility for representativeness of samples or contamination problems during sampling, sample storage and shipping to the respective laboratories. These guidelines should be followed as precisely as possible and any deviations from the guidelines must be reported in the sampling protocols.
The purpose of this study is a first quantitative screening of perfluorooctyl sulfonate (PFOS), perfluorohexyl sulfonate (PFHS), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA) and perfluorooctyl sulfonamide (PFOSA) (Figure 1) contaminants in water (freshwater, seawater, effluents and rainwater), sediment, sludge, fish liver and seal liver from the Nordic countries (i.e. Denmark, Sweden, Finland, Norway, Iceland and Faeroe Islands). This will allow to assess the existing level of contamination (spatial distribution monitoring) possibly indicating regional differences. The spatial distribution monitoring programme will enable to determine the representativeness of the monitoring sites with regard to spatial variability in contaminant concentrations and will give information about the ubiquity of PFAS distribution in the Nordic countries.
2. Perfluoroalkylated substances (PFAS) Perfluoroalkylated substances (PFAS) have been manufactured for more than 50 years. They are used for a large number of industrial applications such as protection of carpets, textiles, leather products as well as in the production of fire-fighting foams, herbicides and insecticides, lubricants, paints, adhesives and acid etching solutions (Kannan et al., 2001; Laikhtman and Rohrer, 1998; RIKZ 2002). About 3250 tons of PFOS-based chemicals were produced by the 3M company (USA) in 2000. Though the manufacturer voluntarily has phased out most of this production volume, similar compounds with long perfluorinated chains, continue to be produced for comparable applications. Compared to hydrocarbons, perfluorinated compounds display exceptional physical properties:
I) They are immiscible with most other liquids. II) Fluoro-organics are non-flammable and non-corrosive. III) Perfluorinated compounds have a very high insulation resistance. The exceptional properties making man-made PFAS so attractive for industrial applications also impose a risk for the environment and ecological systems. The strength of the carbon-fluorine bond makes these compounds very persistent towards degradation. Recent publications indicated that PFAS and especially PFOS are widely distributed over the northern hemisphere, including remote areas such as the Arctic (Giesy and Kannan, 2001; Hansen et al., 2002; Kannan et al., 2001).
3. General sampling strategy Sampling should be performed in accordance with general sampling strategies for chemical trace analysis. In case of questions about the practicability of procedures or
84
84
usability of special material and equipment NILU must be contacted (Roland Kallenborn, phone: +47 77 75 03 86, e-mail: [email protected]; Urs Berger, phone: +47 77 75 03 85, e-mail: [email protected]). The sampling strategy should take into account the specific objectives of the monitoring programme, including the quantitative objectives. Natural variability within the samples should be reduced by an appropriate sampling design. The sampling strategy is an intrinsic component of the data, and may limit their use and interpretation.
3.1 Sampling site selection / representative sampling Nothing is known about the homogeneity of the distribution of PFAS in the Nordic environment, since no comprehensive screening has been performed in these countries yet. Therefore, it is difficult to give recommendations about the choice of representative sampling sites. Only a relatively high number of samples taken from as many different places as possible can overcome this challenge and will add to our knowledge about distribution patterns of PFAS in the environment. The detailed sampling site selection lies within the responsibility of the sampling institutes. Sampling sites must be indicated on the sampling protocols as accurate as possible (preferably with latitude/longitude data). 3.2 Homogeneity of samples Primary sample amounts should be as large as feasible and homogenised on site to yield sub-samples of at least the required volume for analysis (see Chapter 5). Larger sample amounts are preferred (and mandatory for some samples of each matrix) to perform laboratory replicate studies. These as well as field replicate studies (see Chapter 7.2) are an integrated part of the quality assurance programme of the planned study.
4. Sampling equipment / risk of contamination All equipment, materials and containers that come in contact with the samples must be rinsed with high-purity water and methanol before use. Fluoropolymeric materials pose a significant risk of contamination with PFAS, especially for PFOA. Equipment made of or containing fluoropolymers such as PTFE (‘Teflon’) or Viton rubber (sealing rings etc.) must not be in direct contact with the samples and should completely be avoided when handling, storing or shipping samples. The target analytes are surface-active compounds. Furthermore, they display very poor solubility both in water and organic solvents. Currently there is no consensus among experts as to whether PFAS adsorb irreversibly to glass surfaces or not. Glass containers should therefore be avoided and replaced by polyethylene (not PTFE!). NILU must be contacted in case of questions about the usability of certain materials in contact with samples (Roland Kallenborn, phone: +4777750386, e-mail: [email protected]; Urs Berger, phone: +4 77750385, e-mail: [email protected]). Samples should be collected in the same containers in which they are cooled/frozen, stored and shipped to the analysing laboratories to avoid losses due to adsorption and change of vessels.
5. Field sampling / required sample amounts 5.1 Water General water sampling strategies for chemical trace analysis should be followed. Sample amounts of 500 – 1000 ml are required to guarantee good detection limits. The sampling depth in lakes and the sea is crucial. Due to their surface-active properties,
85 of 107
PFAS might form a layer on water surfaces or preferably bind to particles. Therefore, it is important to state if samples were collected on the water surface, several meters from surface and bottom or close to the bottom. Water samples should be kept cool (but not frozen) and in darkness in order to avoid degradation of analytes. 5.2 Sediment General sediment sampling strategies for chemical trace analysis should be followed. PFAS are expected to be distributed unevenly in sediments. Therefore, as much sediment as feasible should be collected over a large area and thoroughly blended before an aliquot is taken. A homogenous aliquot of approx. 100 g should be sent cooled to the analysing laboratory. 5.3 Sludge
PFAS are expected to be distributed unevenly in sludge. Therefore, as much sludge as feasible should be collected and thoroughly mixed before an aliquot is taken. A homogenous aliquot of approx. 100 g should be sent cooled to the analysing laboratory. 5.4 Fish and seal liver
Fish and seals should be dissected immediately after collection and liver samples should be removed and frozen. The fish liver samples should be pooled samples prepared from at least 10 individuals of similar size (weight and length). The weight of the total pooled sample should in any case exceed 5 g. Fish caught during the non-breeding season is preferred over fish from the breeding period. The seal liver samples should preferably be prepared as pooled samples from several individuals if this is feasible, since seals have shown a large variability in PFAS concentrations. A minimum of 5 g seal liver sample is requested for analysis. All biological samples must be frozen immediately after catch and preparation.
6. Storage and shipping of samples The target analytes are generally very stable towards degradation. Therefore, the time between sampling and analysis is not expected to be crucial in this study. However, little is known about enzymatic degradation, therefore all samples should be kept cold and especially biological samples should be frozen (-18 °C) immediately after collection.
All samples must be collected, stored and shipped in clean, fluoropolymer-free containers, preferably polyethylene vessels should be used (see Chapter 4). Containers should not be changed from the moment of sampling or preparation to the arrival at the analytical laboratories. 6.1 Water, sediment and sludge samples Water, sediment and sludge samples must be cooled (< 4 °C) after collection. Water samples should not be frozen. All samples should be stored and shipped at low temperatures and in darkness. They must be clearly and unmistakably marked with a sample name and sent together with their sampling protocols by an express delivery service (TNT, DHL, EMS or similar) to the following address: Norwegian Institute for Air Research (NILU) , Instituttveien 18 ,NO-2027 Kjeller Norway
86
86
To assure that samples reach the destination within short time (usually within the same day), they should be sent early in the morning and not on a Friday (preferably Monday to Wednesday). When sending the samples a notice including the airway bill number (AWB) of the package must be sent to NILU to the following fax number: ++47 6389 8050 addressed to Arve Bjerke and Ellen Katrin Enge. The delivery should be marked with “samples NMR-PFAS study” to avoid unnecessary delay during the registration procedure at the analysing institute.
6.2 Fish and seal liver samples Biological samples (fish and seal liver) must be kept frozen during storage (< -18 °C). During transportation, an insulated box should be used to ensure that the temperature does not exceed thawing temperature (< 0 °C). All samples must be clearly and unmistakably marked with a sample name and sent together with their sampling protocols by an express delivery service (TNT, DHL, EMS or similar) to the following address: Ulf Järnberg : Institute of Applied Environmental Research (ITM) , Stockholm University SE-106 91 Stockholm , Sweden
To assure that samples reach the destination within short time (usually within the same day), they should be sent early in the morning and not on a Friday (preferably Monday to Wednesday). When sending the samples a notice including the airway bill number (AWB) of the package must be sent to ITM to the following fax number: ++46 8 674 7637 addressed to Ulf Järnberg. The delivery should be marked with “samples NMR-PFAS study” to avoid unnecessary delay during the registration procedure at the analysing institute.
7. Sampling quality assurance Quality assurance is a management scheme required to ensure the consistent delivery of quality controlled data. To minimise the risk of contamination or the loss of analytes (and so to avoid the generation of false data) all procedures including sampling, storage and shipping must be evaluated and controlled. This is partly done by analysing field blanks and replicates. Furthermore, sample replicates allow to assess the precision of the complete analytical method. Field blank and replicate samples will be analysed and reported in addition to the 100 samples without additional costs. The steering group of the Nordic Chemicals Group decides where blank and duplicate samples will be taken and is responsible for the distribution of these tasks (see also 7.1 and 7.2). 7.1 Field blanks Field blanks must be taken parallel to the samples for all matrices at five different sampling sites. For water, sediment and sludge the blank surrogate matrix must be treated in exactly the same way as the real samples from the moment of sampling in the field. As surrogate matrix for water analysis high-purity water should be used, for sediment analysis high-purity silica should be employed. A slurry mixture of high-purity water and silica with similar water content as the sludge samples (i.e. only silica if desiccated sludge is sampled) should be used as blank surrogate matrix for sludge analysis. For fish and seal liver the blank surrogate matrix must be treated in exactly the same way as the real samples from the moment of dissection of the animal and isolation of the liver. Triolin (= 1,2,3-tri-(cis-9-octadecenoyl) glycerol) should be used as liver surrogate matrix for field blank experiments.
87 of 107
7.2 Field replicates Additionally to the blank and real samples, duplicate samples must be taken for all matrices at a minimum of three different sampling sites (preferably three different countries). These samples will be used to assess the repeatability and representativeness of the sampling procedures. The duplicates must in each case be two independent samples simultaneously taken from the same site and not two aliquots from the same primary sample homogenate. It has to be indicated on the sampling protocols which samples belong together as field duplicates.
• Hansen K.J., Johnson H.O., Eldridge J.S., Butenhoff J.L., Dick L.A. (2002) Environ. Sci. Technol. 36, 1681-1685
• Kannan K., Koistinen J., Beckmen K., Evans T., Gorzelany J.F., Hansen K.J., Jones P.D., Helle E., Nyman M., Giesy J.P. (2001) Environ. Sci. Technol. 35, 1593-1598
• Laikhtman M., Rohrer J.S. (1998) J. Chrom. A 822, 321-325
• RIKZ (2002) Perfluoroalkylated substances, aquatic environmental assessment; report RIKZ/2002.043, Hekster F.M., de Voogt P., Pijnenburg A.M.C.M., Laane R.W.P.M. (authors), Directoraat-Generaal Rijkswaterstaat, Rijksinstituut voor Kust en Zee, Den Haag, The Netherlands
88
88
89 of 107
90
90
91 of 107
92
92
93 of 107
94
94
Nordic cooperation on screening of perfluorinated substances
95 of 107
Supplementary information: Sampling and sample handling manual from 04.07.2003 Sludge samples: Recent literature suggested that PFOS and PFOA could be anaerobically biodegraded in activated sludge (Schröder, 2003). This problem might occur in wet sludge samples kept at ambient or higher temperatures. To avoid the possibility of degradation of analytes the following recommendations should be followed when sampling, storing and shipping sludge samples: • Drained sludge samples (dry samples) are preferred over wet samples. Analysing only dry sludge would also guarantee the comparability between sludge samples from different countries. • Samples should be frozen immediately after sampling and kept frozen during storage and shipping. • If dry or drained sludge is not available, wet sludge samples with high water content can be deactivated by adding 0.5 % (weight) of formic acid (HCOOH). This treatment must be clearly indicated on the sampling protocol. These samples should be cooled but not frozen (compare also water samples). Reference: Schröder H.F. (2003) J. Chrom. A (2003), 1020: 131-151
96
96
Appendix 3: Concentration list
Sediment-sludge [pg/g ww]
Sample no Type PFOSA PFHxS PFOS PFHxA PFOA PFNA SUM
